The 3-part view of power generation

This is a guest post on my request from DoDo, contributing editor at European Tribune, who works in the railway sector.

In technical English (and many other languages), electricity generation is commonly divided into two basic load regimes: base load and peak load. However, other languages recognise three basic regimes (for example German: Grundlast - Mittellast - Spitzenlast), and this division also appears in English in the usage of some international bodies (for example ENS).

In this article, I want to demonstrate why the 3-part view makes more sense, use it to show the place of exports and renewables in the power mix, and say a few words about the prospects of de-carbonising electricity generation.

Base, intermediate & peak load

In the common view of technical English, in the conventional power plant mix, there are

  1. base load plants, which run at constant power to meet the minimum of demand; and
  2. peak load (peaker) plants, which are supposed to meet the variable part of demand, that is, operate in a load-following way.

Focusing on the variation with the highest amplitude, the diurnal one, the above view can be represented by the following diagram:

In reality, the variability of demand can be predicted rather well: the daily, weekly, annual fluctuations are very pronounced. Thus, the bulk of load-following can be planned long ahead, making it a scheduled form of operation. For the power plant operator, scheduled operation also means that the plant's average load factor, even if well short of 100%, is rather stable and predictable. And this is how we get to the simplest version of the 3-part view of electricity generation:

  1. Baseload: plants operated at constant power output
  2. Intermediate load: plants operated with slow variation in power output on regular schedule to follow expected variation in demand
  3. Peak load: plants operated with fast variation, responding to minute peaks in demand above or below the pre-planned part of supply

Slow-moving intermediate load plants tend to have a non-zero minimum power, and negative deviations from predicted demand are best avoided by giving peaker plants a buffer, too; so both load-following regimes are rarely zero. Graphical representation:

The above picture still contains an over-simplification that hides two basic balancing functions of the system.

  • From time to time, conventional baseload plants (typically in the GW range) need to be shut down: for refuelling, or regular maintenance, or upgrades. This is a scheduled change in baseload, hence, the task of compensation falls on intermediate load, which must have reserves of the proper magnitude.
  • The power of conventional baseload plants (and not just baseload) can go away in an unplanned way, too, and that rather fast: in case of an accident or power line damage. In this case, compensation is a duty of peak load, which again must have reserves of the proper magnitude.

To include the above two functions, one has to recognise that baseload is not constant, which calls for a re-worded definition.

  1. Baseload: plants operated at maximum whenever possible
  2. Intermediate load: plants operated with slow variation in power output on regular schedule to cover the gap between expected demand and expected baseload
  3. Peak load: plants operated with fast variation, responding to minute variation in both demand and the scheduled part of supply

Graphical representation:

The reason power plants of different types are predominantly used in one of the three regimes is mostly economic:

  • If the majority of the lifetime costs of a power plant are upfront investment costs (that is mostly the construction costs), or if the plant depends on fuel coming from an open-cast mine (high investment costs in mining itself), then the unit costs of electricity produced will be the lower the more the plant is operated (resp. the more the mining equipment is operated). Hence the owner will want to operate it at maximum whenever possible (the very definition of baseload as per above), and technologies for variable output won't even be developed much.
  • If investment, operating and fuel costs are all in the same magnitude, then the plant owner will want to at least operate the plant in a regular way, when at least the average output is steady. This leads to intermediate power.
  • If fuel costs dominate, the plant will generate profit whenever on-line, and regularity is not a factor. This is ideal for peaker plants.

With the above, my description of the 3-part view of conventional power generation is complete. Of course, reality is more complex: power scheduling may be subdivided into lots of sub-regimes (ones with 6-hour, 3-hour, one-hour, half-hour, 15-minute, 5-minute tranches, each scheduled ahead for different lengths of time), individual power plants might supply multiple regimes (as gas often does), and there is energy storage. But, these three regimes already describe most basic functions of the system. So now we can explore how some less conventional modes of power generation fit into the picture.

Cross-border flows

It is widely known that some European countries are significant exporters of electricity, while others are significant importers. What is less widely known is that cross-border flows are exploited for balancing, too.

Let me demonstrate this on the example of France. Grid operator RTE publishes load statistics online, mostly in hourly detail. On the diagram below, I averaged one week's worth of data:

(Note on choice of a shorter time period to average: before 1 July 2010, RTE's published data did not separate coal and gas.)

It is already well visible that cross-border flows, which are a scheduled form of power, have a significant contribution to variation. For a better visualisation of the relative contributions to variation, I subtracted the minimum from each curve and stacked the residuals:

One way to interpret the above is that France is simultaneously exporting baseload and importing intermediate load. Or, we could say that France's electricity sector is not a closed system, the whole system includes the importers. Pointedly, we could say that France needs flexible fossil fuel plants in Italy and elsewhere to maintain a high grid penetration of relatively inflexible nuclear energy.

Intermittent renewables

In recent years, the number one contributor of new capacity in Europe has been wind power. Even more recently, solar power, more precisely photovoltaics, gained significance: though still rather expensive, prices began to fall rapidly and annual installations are now in the gigawatt range. (In Germany, last year the electricity production of photovoltaics surpassed 1% of net consumption, about the level of wind power in the USA.)

In the lifetime costs of both wind power and photovoltaics, fixed, up-front investment costs dominate. Hence, owners will want to operate them without limitation whenever possible. Indeed the preferred support mechanism for renewables across Europe, that of feed-in laws, obligates utilities to purchase all production at fixed feed-in rates, effectively putting them at the start of the merit order. Thus, while it is common to treat wind and solar as separate from the two resp. three classic load regimes, by the final definition above, feed-in law covered wind and solar are de facto operated as part of baseload.

The one difference compared to conventional baseload is intermittency: for both wind power and photovoltaics, power output depends on weather, time of day, and season. When turbines (resp. panels) are distributed over a grid spread out over a larger geographical area, the weather-related intermittency can be reduced, but doesn't go away.

However, in countries and regions with high wind penetration, grid operators use wind prediction methods that are by now sufficiently precise 24 hours, 36 hours ahead to enable the bulk of the balancing to be scheduled. Solar power can be predicted efficiently, too. Hence, intermittency gives most of the balancing job to intermediate power, where significant reserve capacities already exist. The residual, the difference of actual and predicted output, has to be balanced by peak load.

As emphasized before, conventional baseload is not constant, either. What differs is the pattern and spectrum: instead of relatively fast stepped changes, there are slower but continuous swings -- similar to the other variation intermediate load has to balance, that of demand. As for peak load, the unpredicted short-term fluctuations are not dissimilar to the other irregularities to balance, such as demand surges or blackouts or power plant accidents.

The connection between intermittent renewables and baseload is also apparent in studies on the feasible maximum grid penetration of wind et al in the future: in these studies, classic baseload disappears (see f.e. the diagrams in a German study), and even if power plant types presently operated as part of baseload are assumed to continue, then in a load-following way (f.e. in another German study on operating a future system with or without nuclear exit [diagrams 4a and 4b]).

De-carbonising intermediate load

While there are mature technologies to de-carbonise baseload production, as discussed in the previous two sections, intermediate power has to remain significant whatever is chosen as baseload, and intermediate power is dominated by fossil fuel. What are the potential substitutes?

One mature technology already in use (see diagrams for France again) is hydroelectric power. However, there are capacity limitations: conventional dams have already tapped much of the resources available outside conservation areas, and those resources are rather uneven geographically. The version employing active energy storage, pumped hydro, needs lots of water, too.

Another renewable technology maturing recently is burning biomass. This technology provides for a more or less direct replacement of coal, gas and oil (and thus both intermediate and peak load provision). However, biofuels bring a host of other problems, above all a low EROEI and a conflict with agriculture over the limited arable land available.

While the prior discussion focused on supply-side management only, demand management has a history, too. For example, in Sweden or France, electrical heating has been advocated because home heating consumption peaks during the night, when other consumption is lower. Ideas for the future include the night-time charging of the batteries of electric cars. However, such demand-side solutions tend to have their limitations due to seasonal variations: for example, electric heating is not used in the summer. Hence, while demand management might reduce the annual total of intermediate power generation (as demonstrated by a comparison of the previously shown graph of diurnal power variation for July with the one for February below), the capacity requirements stay the same.

In the previous section, I considered wind and solar intermittency separately. However, on one hand, there is some negative correlation between the two: when the sky is cloudy, winds tend to be stronger, and solar power peaks in the summer while wind tends to peak in the winter. On the other hand, solar power peaks during the day, while wind tends to peak in the afternoon. Thus, it is possible to combine appropriate capacities of wind and solar so that the combined power output will fluctuate around an average that resembles the demand curve. This is natural balancing. Like demand management, this is more a way to reduce the volume of intermediate power than its required capacity.

Solar thermal power generation, when using a heat transfer medium that can be stored, can have a regulated output resembling the demand curve, too. But land area suitable for large-scale solar thermal is geographically limited.

Though there are a host of potential maintenance issues, it is technically possible to operate nuclear plants in a load-following way, at least in the intermediate regime. However, due to the earlier mentioned economic aspects, this has been done just with some aged plants, with a few that already have amortised investment costs. The EPR is the first full-scale model capable of large-amplitude power variation on a daily basis by design, but it would need government commitment to operate one that way, and years of operational experience to confirm that maintenance won't become an issue.

Classic geothermal power, tapping hot water reserves, is a mature technology, but limited to a few geographical locations, and not even renewable if water reserves are drawn excessively. In contrast, stimulated or hot-dry-rock (HDR) geothermal, which pumps water into deep dry boreholes, is one of the renewables with potential in the order of magnitude of consumption. This technology allows variable output, and has significant operating costs (needs water, and maintenance of parts under pressure and thermal strain), and thus it is economically suited for intermediate load provision. However, the technology is still at its infancy, and pilot projects indicate that boreholes bring water reservoir and earthquake risks.

Beyond pumped hydro, another energy storage technology is pumped air: cavities like disused mines can be used as reservoirs to pump air in and then let it out via turbines. This technology is at its infancy, too.

There is the possibility of distributed energy storage, too: grid-connected flywheels, batteries, capacitors, fuel cells can be operated in a grid-supporting way ("smart grid"), and such de-centralisation even increases local security of supply. However, these methods of storage tend to have low RTE and EROEI.

In conclusion, I would say that there is no silver bullet, de-carbonising intermediate load will probably need most of the above at the same time.


The author wishes to thank European Tribune and TOD:E editor Jerome a Paris for contributing source material and criticism.

Very good treatment.

Something I'd note is that compressed-air energy storage (CAES) is very small at the moment, but uses standard turbomachinery which can be built in large quantities in existing plants.  This gives it potential for extremely rapid growth.  One fledgling effort is the Iowa Stored Energy Park.  CAES appears well-suited to supply both peaking and intermediate load.

Also AA-CAES could share thermal storage with a solar thermal plant, which could piggyback onto a combined cycle gas plant giving you solar assisted combined cycle with thermal and compressed air storage which would offer a lot of flexibility.

Also we seem to forget the point that fossil fuels are a store of energy as much as a source of energy

From what I understand, the difficulty is with heat: compressed air heats up, needs to be cooled, but when it is let out across the turbine, it has to be heated up again to prevent moisture problems. There are research projects to do this with heat storage instead of heat loss and gas re-heating, but the ISEP seems to be conventional. BTW, I can't find anywhere the capacity of the ISEP, do you know what will be its maximum in MW? (I guess much less than the planned Ohio plant.)

Depending on the moisture content and compression moisture will come out of the air when it is compressed and when cooled. Air can only carry a fixed partial pressure of water vapour at a particular temperature. Once you expand the air the amount of water vapour is the same but the volume of air is greater so the PP will fall. It can only release that water vapour if the temperature falls low enough that the PP of water vapour falls below that of the PP of water vapour in the air. Water would need to be collected from the cooling, compressed air and storage caverns.

This is why we dry air before compression into scuba tanks to prevent water being left in the tanks. Also, run a compressor until the storage tank is at full pressure.Turn off the pump and let the tank cool (still under pressure). Open the drain valve, at the bottom (carefully as you are under pressure), water will run out.


Depending on the moisture content and compression moisture will come out of the air when it is compressed and when cooled.

That is true if you allow it to cool down. If you don't let it cool, a process that physical chemists would call adiabatic, the increase in temperature increases the dewpoint by much more than the higher density of H2O molecules (per unit volume) increases the absolute humidity. So compressed (but still hot) air is dry air. Of course if you store it in an uninsulated conatiner, it will lose heat to the container (and the efficiency of the energy storage (energy out/energy in) will be low.

If one is just trying to cover the occasional blip in demand, system efficiency may not be a big driver of your economics. But, if you want to store a significant fraction of your power it is important.

That is why I put the 'Depends' in there.


I can't find anywhere the capacity of the ISEP, do you know what will be its maximum in MW?

Their webmaster does need a swift kick for having so little on the site, doesn't he?  Sandia's presentation says 268 MW.

(I guess much less than the planned Ohio plant.)

ISEPA doesn't have a nice, big limestone cavern to play with and has to make do with blowing bubbles in an isolated aquifer.  I wish the Ohio effort well; it's going to increase the regional electric storage by about 150% over the Ludington pumped storage plant alone.

As for heat of compression, it's relatively cheap to store if you just make the thing big.  A pressure chamber full of small stones or other media can store a lot of heat; you blow hot air through it in one direction on the way to storage, and let cool stored air flow back the other way when extracting energy.  I suspect this is the general scheme General Compression is using.

Thanks for the details!

Below I'll show another diagram, which shows an actual, extreme case of load balancing, and demonstrates the conflict between FIT-supported intermittent renewables and conventional baseload.

The diagram is from a a German Economic Ministry study into the phenomenon of negative electricity stock market prices. Spot prices (red curve on the diagram) turn negative when a producer would rather pay for the consumption of electricity generated than power down his plant. Here is one such event on a one-week long diagram, with load curves also shown:

Load curves:
Light blue - wind
Yellow - nuclear
Brown - low-grade coal (from open-cast mines)
Black - high-grade coal
Orange - gas
(Hydro & pumped storage & biofuel & rest not shown)

We see that during a big peak of electricity from wind,

  • the main providers of intermediate power (high-grade coal and gas) are powered down,
  • some low-grade coal plants (part of baseload) are also powered down in phase,
  • * when wind output peak and demand minimum coincide, even a nuclear plant (also baseload) is shut down.

The ministry claims that negative spot market prices are a problem, and, as one can expect for a ministry led by a member of the free-market-advocating FDP party, sees the solution in moving wind power from the feed-in law to the open market -- what that means is that some wind farm owners would have to power down in times of high wind and low demand for the benefit of low-grade coal and nuclear power plant operators. A different conclusion to draw would be that some low-grade coal or nuclear plants are now obsolete, at least when operated as baseload.

That's a very selective conclusion set. I'd more likely conclude that the value of subsidizing the build of additional new wind generation needs dramatic re-evaluation while solar-thermal with thermal storage needs to be researched and supported.

Why would you want to stop the expansion of wind power?

To avoid wasting all our economic resources on a dead-end venture.

Now that wasn't very detailed...

That looks like a really good case for something like CAES.  Buy at near-zero prices and sell back to the grid a few hours or days later.  Even better, make most of your money from spinning reserve and fast-response generation; that will take more fuel-intensive generation out of those segments.

Slightly off topic, what is the cost of electricity in France? I ask because I am curious about the cost on nuclear power. And for the most part nuclear power only occurs with state backing. Since France is primarily powered by nukes I assume that the price of electricity in France is a good proxy for the true cost of nuclear power.


PS I don't want to launch a heated debate about the merits of nuclear power. I just want a rough idea of the cost.

IIRC Jerome a Paris argued in the past with data that French nuclear electricity is indeed cheap. However, there are problems with this proxy idea:

(1) The composition of baseload might or might not limit price hikes, but the price of electricity in a market environment is determined by the most expensive producer needed to meet demand (this is the merit order effect) -- which will be a peaker plant with relatively pricey electricity.

(2) The total price of nuclear is highly dependent on some uncertain variables that may be subject to state guarantees (reactor lifetime) or become externalities (costs of disposal).

The total price of nuclear is highly dependent on financing cost, which is definitely highly subject to state guarantees.

It's a good question. A French reporter told me that the French government plays many games with the official cost of nuclear energy to hide the true, higher cost. But I didn't ask for more detail.

I have lived off the grid for extended times in the field and through hurricanes. There really is only a few things you need here in the near tropics of Bama in the summer.
1. Cellphone. Communication is vital.
2. Stocked ice chest. If you can't have a cold beer or food it really is a disaster.
3. A fan and good ventilation.
4. Minimal lighting.
5. Radio.
6. Water for the toilet and to drink.
I bathed in the river and I could ride my bike to the store or work.

Put that into the average usage chart and I come up less than 10KWh per day. It might be less than 5. Forget get the rich and the comfortable, just think if only the fringe and the poor lived like this. Power needs would drop significantly.

I just read my electricity metre. Family of 4, average sized UK house. Daily consumption 4.7KWh.

That includes cooking, computers, phones, electronics, lighting, freezer etc.

Not much call for air con round here, and heating is by gas and solar.

We joined a local survey on electricity usage. We were the second least profligate. The winning family used 3Kwh.

[edit] the etc. includes (electric water heating) washing machine, kitchen appliances, kettle, vacuum cleaner, tv, games machines, radios occasional use of tumble dryer, machine tools, dehumidifier, and more.

We have most of the gadgets. we use them sparingly.

Family of 3, average UK sized house. . . 1.6 KWh per day.

We have a pretty 'normal' lifestyle (computers, tools etc) apart from getting rid of the fridge. (Was 2.5 KWh before that.)

It seems crazy to me that we have got to the stage now where if the grid goes down, people start to die! Really scary the way we design our cities and houses . . .


Just paid my bill. Consumption went up a touch, bill went down. Government subsidises more in the summer to allow for people to keep cool. 5.29 kWh much of which is the fridge freezer. There is a government scheme to replace old ones with new ones but mine isn't quite old enough to come under the scheme despite using a lot more electricity than modern ones :( I really must get that solar panel put together and get some solar lighting going.


25.77 kWh average per day over the year for seven people in a poorly insulated house in Boston. Each of us has a computer running basically non-stop (engineers and computer programmers, so we're all working on them often). Refrigerator and chest freezer. CFL lights everywhere, and usually off due to extensive natural lighting. Electric stove, which probably combined with the refrigerator are the primary power users. AC during the summer, but doesn't seem to significantly effect the overall power usage.

It would seem to me that as wind power is ramped up, more and more of it will end up in the cross-border flows category. In this category, I would expect that feed-in tariffs would no longer be applicable (or if they are paid to the producer, the funds that are received for the power that is actually sold cross-border would be much lower). At some point, there will no doubt be wind power that is sold for a negative amount, because it is truly excess.

So as wind power ramps up, there will be more and more of a funding issue for someone (not sure who ends up with the shortfall), because of the low prices the unwanted power gets when it is sold across borders. in order to keep the system financially solvent, it seems like one would have to keep ramping up the feed-in tariffs.

Has anyone done studies of how the amount of cross border power sold increases with an increased number of turbines? I would expect with the first few turbines, this would be relatively little of an issue. But as the percentage penetration goes up, it would be more and more of a problem.

All of the averaging is nice in theory, but it would seem to me with borders in the way, it won't really work as planned. The seller will very often get a low price for the locally unwanted wind, regardless.

I imagine cross-border flows would bring losses only if they are one-way, or if the imports in times of low wind and high consumption are more expensive. But, from the little I know, neither is the case in the only place where this happens, Denmark: the balance is with Swedish and Norwegian hydro.

Everything I have heard says that Denmark comes out fairly much behind in selling to Norway and other countries. For example, Wikipedia says:

While wind power accounts for almost 20% of the electricity generated in Denmark, it covers only 10–14% of the country's consumption. Power in excess of immediate demand is exported to Germany, Norway, and Sweden. The latter two have considerable hydropower resources, which can rapidly reduce their generation whenever wind farms are generating surplus power, saving water for later. In effect, this is a way for northern Europe to store wind power until it is needed – an opportunity which is not generally available for wind power generators in other regions.[15][16] The benefit of this goes to Denmark's neighbours; when Denmark's wind farms are exporting power, it is sold at the spot market price, which sometimes falls to near zero – or even lower.[17]

This is what I would expect, based on logic. And if Denmark adds more production, it will disproportionately be exported, making the problem worse.

If Denmark guarantees a feed in tariff to domestic wind energy producers, Denmark's electricity customers will end up paying what is in effect a bigger and bigger subsidy for wind they pay a high price for, that gets sold abroad for low prices. We know residential electricity prices are very high in Denmark (compared to other European counties and the US). This may be the primary reason why.

If the amount exported could remain small, and in line with what is imported, it might not be as much of an issue.

Yawn. The misleading arguments in the quoted paragraph on Wikipedia seem to be lifted from the infamous (IER-commissioned) study by the think-tank CEPOS. The last sentence says nothing about the balance of electricity sold and bought back, and Nordic hydropower is very cheap, too.


Can I suggest that you edit the Wikipedia article? Heck, that's the whole point of Wikipedia - when we see errors, we can correct them.

When I have spare time :-) (I do edit Wikipedia from time to time, but to replace a bad text with sourced material is a lot of work.)

Well, sourced material is certainly the best thing. OTOH, if you're confident of the information I'd enter it now, and return with the sources when you can. The perfect is the enemy of the good...

Well, I have been involved in some mini WikiWars when I corrected some contentious issue unsourced, and it was re-edited within hours...

Now that you mention it, I have too. That can be frustrating.

Well, I added your source to the wikipedia entry that Gail referenced - perhaps that will help. I added it as a link, rather than a formal reference. We'll see what happens...

The link was turned a reference :-) I have now updated the Wiki page, too, by expanding on the counter-argument, and also updating the statistics in the table. However, I cut the reference to France: it's not that France imports by day, but that exports are reduced by day (and at the moment I can't offer a better source than this article of mine).

Now I have to run, will check the 32 other new comments during the European night later...

Having checked the debunking of the CEPOS study, I find even the notion that it is wind power that gets exported is erroneous. The linked paper first reproduces the analysis of the CEPOS study: calculating correlation of exports with power generation, but this time, for both wind and conventional generation (both relatively low). Then the paper does causal analysis and shows evidence that Denmark's large thermal power stations chose to export when Nordic spot market prices were high -- the merit order effect at work, which is then explained in exhaustive detail. They estimate the exports truly caused by wind over-supply (taking into account the non-zero minimum power of CHP plants) at a mere 1% of wind production. Then an analysis of the overall merit order price reduction effect follows.

Something the above study doesn't do either is calculate total costs. With much struggle (Excel formats...) and little usable result, I tried to do so myself.

I downloaded the hourly spot prices and hourly scheduled flows for all of 2008 from the Nord Pool Spot site. The problem is: a strong component of the flows across Denmark's borders is transit from Sweden & Norway to Germany, but the bulk of that exits Denmark through non-NPS channels. Hence, while all of Denmark had net imports of 1.54 TWh that year, in NPS, the net imports were 8.29 TWh; and with Sweden and Norway only (excluding Kontek, the NPS connection to Germany), 9.13 TWh.

So, I thought maybe the avergage price of imports/exports would show something significant, but there isn't much:

  • Average NPS spot price: 44.72 €/MWh
  • Average price Denmark paid for net scheduled flows on NPS: 44.31 €/MWh
  • Average price for which Denmark sold the sum of hourly net exports on NPS: 43.81 €/MWh
  • Average price Denmark paid for the sum of hourly net imports on NPS: 44.24 €/MWh
  • Average price Denmark paid for net scheduled flows on NPS, wrt Sweden & Norway only: 45.07 €/MWh
  • Average price for which Denmark sold the sum of hourly net exports on NPS, wrt Sweden & Norway only: 43.40 €/MWh
  • Average price Denmark paid for the sum of hourly net imports on NPS, wrt Sweden & Norway only: 44.84 €/MWh
  • Average price for which Denmark sold net scheduled flows on NPS, wrt Germany only: 52.60 €/MWh
  • Average price for which Denmark sold the sum of hourly net exports on NPS, wrt Germany only: 47.75 €/MWh
  • Average price Denmark paid for the sum of hourly net imports on NPS, wrt Germany only: 43.97 €/MWh

The only obvious trend here is that Kontek trade with Germany was to the Nordic countries' benefit. For a more meaningful analysis that is less skewed by transit, I'd need the hourly data on the non-NPS exports to Germany, and maybe I should download and process the 2009 data, too...

By the way, the EU Commission's Energy Commissioner, Günther Oettinger recently brought up that there should be Europe-wide feed-in laws. Methinks that's politically impossible right now, with the fundamental differences in the philosophies of say the Spanish and German feed-in laws, but an interesting idea on the long run.

The currently seen negative pricing will probably only be short term, until storage systems ramp up. Sodium sulfur and sodium iron chloride batteries are potentially cheap, environmentally benign, and efficient. The economic incentive to buy electricity at low or zero cost and sell it within the same day at peak rates will ensure that utility battery systems are incorporated in the grid. Eventually the variation in electricity price will be bounded by the efficiency of the storage systems used; low price = high price * efficiency. For a 75% storage efficient system that means variations of no more than 5 cents/kWh.

The energy payback time of a modern wind turbine is about 6 months. That of a concentrating photovoltaic system is 8-12 months, while that of planned nuclear plants may be about a year. All of these options yield high enough EROEI to allow rapid phase-in without energy cannibalism from depleting fossil fuels. Its past time to accept arguments against wind and solar as the FUD that it is and just get on with it!

The currently seen negative pricing will probably only be short term

Absolutely. Various forms of demand will move to night time to take advantage of cheap power, chief among them EV charging.

I noted in a recent local article that wind power has its own problems when integrated with the grid:

Even though the Pacific Northwest gets a large portion of its power from hydro, it's possible to throttle back hydro only so far.

What emerges from this is that we need a better power grid--perhaps that should be given highest priority. And energy storage will continue to be a big problem.

The BPA is exaggerating their problems with wind.

Here's a map of their wind farms: , from .
We see that about 5% is about 70 miles from the rest, about 7% are about 40 miles away, but most are within a circle of a radius of about 20 miles.

This wind resource isn't that large, and it's in a very small area, so it makes sense that it would have a fair amount of variation.

I am a BPA employee. I don't think BPA has exaggerated their problems.

The resource is 2000MW of wind capacity. The peak power load in BPA's balancing authority is 15000MW. It averages more around 6000MW (you can see real-time data here:

BPA markets power from federal dams on the Columbia and one nuke plant at Hanford. The dams operate within environmental limits to protect endangered salmon. They must spill water over the top of the dams to help the smolt get to the ocean during the spring. But they cannot spill too much, as excess dissolved nitrogen is problematic for the fish. The river flow planning is hampered by limited storage below Grand Coulee. In late May, there was an unusually high stream flow and hydro operations were constrained. There was very limited margin to back off of generation to accommodate the high wind generation. Unfortunately, the weather was relatively mild, so there was not a lot of generation demand, though I do recall the AC and DC interties being heavily loaded. The power dispatchers called for a reduction of wind generation to avoid an environmental violation. I believe there are pre-existing agreements to do this.

I think BPA is doing a very good job of accommodating wind in its balancing authority. But there is this idea in the agency that BPA should not accommodate more wind generation destined for other balancing authorities if it increases costs for the ratepayers of the Pacific Northwest. They are working on creating contracts to allow variable transmission contracts out of the BPA balancing authority. There are also very real limits to power flow out of the Mid-Columbia grid to Portland and California.

Anyway, I would be interested in knowing in what way you feel the problems are being exaggerated.


Jeff Barton

That's very interesting that BPA has a wind resource rated at about 1/3 of average consumption.  What's the capacity factor?

Having to turn down production during low-demand periods could be seen as a feature; it means that capacity is getting to about the optimum point for the local grid.  Being able to export more power would change that point.  The vision of a heavily wind-supported grid includes widespread interconnections to take advantage of regional variations, and that fits right in.

I don't know the capacity factor of the wind connected to the BPA grid, if I understand your question. My guess in the 20-30% range, based on the limited exposure I have to the power side. You might find the answer in the BPA wind site (see my post below).

In the past if BPA wanted to sell the wind generated in its balancing authority to other entities, it had to firm up the wind. The hydro generators are good at this, but there are limits to this capability. With limited geographic diversity, all the windfarms ramp up and down more or less at the same time. I think there are efforts ongoing to change regulations governing the WECC that were designed for a fossil fuel/hydro era. There are also some technical issues (power flow stability/reliability and transmission congestion) that need to be addressed to integrate intermittent resources over a large area. Still, I think things are heading more or less toward a lot more wind integrated into the power grid. I can almost see a future where the different renewable energy zones in the WECC ( will be linked with AC or HVDC lines, sharing generation diversity from Alberta and BC to New Maxico and Arizona. It almost makes me optimistic. Almost.

Jeff Barton

Thanks for the information.

The exaggeration I'm referring to ocurred in news articles, so it may have come from the journalists. I'd have to look back at the articles I read to identify what specifically made me say this. In general, there was a tone of skepticism towards wind that I felt was unwarranted.

I don't see any sign of such skepticism in your response. I'll be more precise in the future, and say something like "popular news articles about BPA's wind resource have exaggerated the problem".

I agree that the news article was a bit over the top.

BPA does have a wind integration team and, from my internal perspective, a strong commitment to gettign as much wind on the grid as is feasible without compromising reliability or violating commercial or regulatory constraints.

As evidence I would offer that:

BPA was early (possibly first) in offering an intermittent generation connection that dropped the $100/MW penalty for generators that significantly deviate from schedule.

BPA was a big player in developing the "Network Open Season" which has gone a long way to allowing BPA to consider costs of interconnection for a kind of pooled group of developers, rather than on a project by project basis. Not my area of expertise, but I understand it helps break the logjam of plants not getting build due to lack of transmission.

BPA is cooperating with other utilities in the WECC to share variability of intermittent resources. The program is just getting online; it is called "ACE Diversity Interchange."

There are other programs and projects, but I am not a wind integration guy.

Still, I think BPA is hitting some technical and/or commercial limits to integrating more wind power that deserve sober consideration.


I assume that the real time data is archived somewhere for later analysis. Is there any way an interested person could get hold of some of the archival data? I would love to do some analysis on this dataset.


Total BPA balancing authority wind data is available at this site:

They seem to have 5-minute data from 2007 onwards, as well as links to installed capacity. The data is in Excel format.


Jeff Barton


Can I suggest a little more discussion of EVs? The article said: " Ideas for the future include the night-time charging of the batteries of electric cars. However, such demand-side solutions tend to have their limitations due to seasonal variations", but EVs don't have a seasonal problem, and they could provide an enormous Demand Management resource. That resource would include both dynamic charging and, eventually, Vehicle to Grid power.

Isn't the amount of driving seasonally dependent, too?

Only slightly. Here's Vehicle Miles Traveled data for the US, in 2009 ( :

JAN 226.3
FEB 219.2
MAR 249
APR 252.9
MAY 260.4
JUN 260.7
JUL 267.7
AUG 263.1
SEP 244.1
OCT 254.7
NOV 239.3
DEC 241.9

Peak: July, only 8% above average.

Interesting, thanks. But could you find the more direct fuel use data? (I couldn't in ten minutes.) Because driving patterns (reflected in fuel efficiency) may fluctuate seasonally, too.

I found these two diagrams for seasonal variation in traffic on highways.

If you look at the link I provided, you'll see that they provide detailed data in spreadsheet format. It breaks down the overall total of VMT into highway, urban, arterial, etc.

Urban driving seems to vary less than rural. Highway driving seems to vary roughly the same as non-highway. I don't see evidence for greater variation in fuel consumption than for VMT.

Keep in mind: we're talking about EVs, whose energy consumption is much less affected by urban vs highway differences.

Keep in mind: we're talking about EVs, whose energy consumption is much less affected by urban vs highway differences.

Whuuuuu? They're wildly affected...just in the opposite direction. They'll go farther at slower speeds than they will at high speed. High speeds kill EV range.

The city MPG for a Prius is only slightly higher than the highway MPG.

No question, if you drive fast enough you can kill EV range. OTOH, a good EV design includes good aerodynamics, which reduces this problem.

Prius is a hybrid and not a pure EV...therefore it trades off between electric at low speed and the gas engine at higher speeds so you wind up with that small plays on the advantages of each motor. A pure EV will drop range dramatically at higher speeds, and not just because of wind resistance.

I just ran across this, and it's a good example:

from Solectria
at 30 mph 86 miles
at 40 mph 65 miles
at 50 mph 48 miles
at 55 mph 38 miles

A pure EV will drop range dramatically at higher speeds, and not just because of wind resistance.

What besides wind resistance would reduce an EV's range at high speeds?

The EV you found has terrible aerodynamics. It's a conversion, for goodness sake, not an OEM like the Leaf or Volt.

AFAIK "good aerodynamics" is a speed-independent variable. Air resistance depends on speed and aerodynamics.


Vehicle energy consumption comes from 1)wind friction, 2) internal drive train friction, 3) braking losses, and 4) wheel and suspension flexing.

EV regenerative braking greatly reduces #3, leaving #1 as the major energy consumer. Good aerodynamics reduces wind friction, making #2 and #4 relatively more important.

EV energy consumption certainly goes up with speed, but the rate at which it rises is far from the simple speed-squared relationship that applies only to the wind friction component.

Finally, to take this back where we started: I see little reason to expect substantial seasonal variation in EV charging demand. Summer VMT rises a little, but not that much. For a long time I expect EREVs like the Volt to dominate, which means that summertime highway vacation travel will have a slightly higher liquid fuel component. Some families will achieve the same effect by using a Leaf for daily driving, and going back to the old gas-burner for the vacation.

Both effects will clip the summertime VMT peak as far as EV charging goes, so there will be very little seasonal variation in charging.

AC Propulsion had a range extending trailer they used for longer trips - not sure if they're using it any more as they're stuffing their cars chock full o' batteries these days and probably don't need it:

Vehicle energy consumption comes from 1)wind friction, 2) internal drive train friction, 3) braking losses, and 4) wheel and suspension flexing.

#2 is greatly reduced in an EV. And at less than highway speed is dominant for a typical ICE powered vehicle. From experience with the Prius, even for a standard hybrid I'd put the savings from the reduction in #2 (whenever the ICE is off) as much greater than the savings for #3.
I would add in rolling losses, primarily flexing of the runbber tire. Perhaps you meant to include that in #4, but it might make sense to split these out. Some folks are working on capturing some of the suspention loses, which are considerable.

Both effects will clip the summertime VMT peak as far as EV charging goes, so there will be very little seasonal variation in charging.

I think the real issue behind the claim that EV charging is a poor match seasonally, has to do with the seasonal supply variation of the renewable electricity, not the seasonal variation of EV demand.

Ah. Well, fortunately (as noted elsewhere) this seasonality can be largely eliminated.

This can be done by pairing wind with solar. Also, careful wind farm site selection can reduce seasonal (as well as diurnal) variance.

"What besides wind resistance would reduce an EV's range at high speeds?"

Drivetrain losses, but primarily the load on the battery pack. Higher speeds require a higher current from the pack, and the more current you push through the pack the less you get back from it.

Small link of Google image search for Single Charge Cruise Range

the more current you push through the pack the less you get back from it.

Ah. That's a much larger problem with lead-acid, than with the latest generation of li-ion batteries, which can put out much more power (kilowatts) without reducing effective energy density (kWhs).

Its just simple ohms law mathematics. You only have so many amp hours in the battery no mater what kind of battery you have. The more amps you take you take the faster the total amp hours are depleted.

You don't have a fixed number of amp hours in a battery. The more amps you draw, the less amp-hours you get.

I do agree with Nick that lead-acid suffers more from this affect than others, but other chemistry is also affected.

For example here is a Trojan 12V Deep cycle gel battery:

From the chart (you can look others up if you wish) you'll note that at a rate which will deplete the battery in 20 hours, you can obtain 77 amp-hours from it. At a rate of draw which will deplete the battery in 5 hours, you only get 66 amp-hours from the battery. Unfortunately, both of these figures are still too high for an EV application which will probably deplete the battery within 1 to 2 hours of driving, which means what you will get back will be even less than either of those two.

How do drivetrain loses scale with speed? If one is using some sort of variable speed transmission I would expect loses per kilometer to go down with speed. I suspect EV powertrain loses are mainly a function of the instantaneous power output. In straight uniform speed driving they are almost the same as speed, but in most realworld situations power will vary considerably -even at near constant speed.

I had a climb a hill this morning with the Prius at 2mph (traffic backup). At such a low speed, I doubt the efficency was very good. Similar to the fact that a human can tire doing an isometric excercise that accomplishes zero work.

It's only a good example if the average urban commuter can drive everywhere at a steady 30mph w/o braking. ;)

The city MPG for a Prius is only slightly higher than the highway MPG.

This is a function of the poor round trip efficiency (circa 60%) of power in/out of the battery, and the fact that few urban drivers try to minimize braking. Supposedly Lithium Ion batteries can acheive 90% round trip efficiency.

I agree we need much more discussion of real-world impacts of EVs. Actual driver behavior, not ideal behavior, will determine their impact on energy consumption and CO2 emissions. We taxpayers are sinking a lot of money into this transition.

The use of air conditioning has a huge impact on Prius mpg. I assume it would have a similar negative impact on EV power usage.

The use of air conditioning has a huge impact on Prius mpg.

The article doesn't provide any numbers - it just makes a broad statement. Have you seen numbers?

The use of air conditioning has a huge impact on Prius mpg.

Over the last 2 weeks of 90 degree temps in chicagoland, my prius avg. mpg dropped from 53.4 to 52.7,
Usually keeping the AC fan setting on LO (square 1) and never above the MD setting (square 4).

The use of air conditioning has a huge impact on Prius mpg.

At highway speed not so much. In my usage the Prius gets more efficient the higher the ambient temperature (at least up to 115F). Once I resort to turning on the AC (and use more of it as the temperature goes up), the milage versus temperature curve then flattens out, i.e. the increased AC demand at higher temps just about matches the increased efficiency. Of course for city traffic, the AC demand stays fixed, while the demand to move the vehicle (per unit time) goes down. Clearly sitting stopped at the light, with the AC running you are consuming considerable power, but going nowhere.

I agree we need much more discussion of real-world impacts of EVs.

Ok, I'll start. Before I do, note that I like EVs and recently was thrilled to learn of the Brammo Empulse.

That said, the real-world impact of EVs is this: not much.

In the U.S., auto sales have nudged up to just under 11 million sales per year. Pretty much none of those are EVs. The Volt is starting with 10,000 per year and the Nissan Leaf with 25,000 per year — peanuts, really. This projection puts EV sales at 400,000 annually by 2020. However, even that number is too high because they haven't banked on a depression occurring between then and now, which will cause sales of all goods and services to decline — especially high-ticket items like cars. I wouldn't be surprised if auto sales in the U.S. decline to 3-5 million per year.

As for electricity, one of the biggest problems we are going to have soon is how to keep the utilities operating as their revenue plummets. We are already beginning to see them adjust electricity rates up because they aren't selling as much as the economy contracts.

Georgia Power requests 10% increase in electric bills as recession cuts expected demand

7% electric rate increase proposed
We Energies blames decline in sales amid recession

If you think there are going to be many EVs on the road in 2020 to make a difference to anything (annual gasoline use, electricity use, pregnancies incurred in the back seat — whatever measure you want to look at), you likely believe some sort of BAU will continue. (An incorrect assumption, in my view.)

Seems a bit circular.

The Original Post is describing practical ways in which we'll have enough energy.

You're suggesting we'll have TEOTWAWKI due to economic depression. Why? Presumably because we won't have enough energy. But...what if we do have enough? Then no depression.

Not circular at all.

First, EVs will represent a minuscule portion of the vehicle fleet by 2020 even if the economy stays in its current limping along condition (i.e. doesn't contract further), so the argument holds no matter what I say next. EVs are just a non-issue any way you look at them for the foreseeable future.

Second, we're certainly headed for our next leg down when the next big oil price shock arrives, which appears to be sometime between now and 2013 or so. Certainly by 2015.

Third, even before then, the world economy is queued up for contraction as this credit bubble continues to burst. This contraction will occur no matter what happens with energy. The stimulus has only delayed the process by a year or so. Further, the G20 are resolved to a future of austerity (including now Obama, to some extent at least) which means that government expenditures are going to decline across the board putting a combined millions of fed, state/province and local government workers out of a job. China's bubble is beginning to pop.

It looks like we're all heading toward a future like what Ireland is in the midst of. Since it will be across the G20, though, I expect the combined effect to be greater.

As Europe’s major economies focus on belt-tightening, they are following the path of Ireland. But the once thriving nation is struggling, with no sign of a rapid turnaround in sight.
Rather than being rewarded for its actions, though, Ireland is being penalized. ... Lacking stimulus money, the Irish economy shrank 7.1 percent last year and remains in recession.

In Ireland, a Picture of the High Cost of Austerity

The next oil shock will just make things worse.

There is no escaping this economic future now, in my view.

The Automatic Earth has just updated its primer. You should read it.

The Big Picture According to TAE - An Updated Primer Guide

We'll have lots of energy soon because as the economy contracts the current infrastructure will be adequate. As I said, the difficulty now will be to keep many of the utilities solvent so that the can continue to deliver power.

EVs will represent a minuscule portion of the vehicle fleet by 2020 even if the economy stays in its current limping along condition

That's a choice. We could ramp up much faster. For instance, we're choosing to ramp up wind power reasonably fast, even though we don't need to. If we continue to ramp up wind power, at some point there will be pushback from legacy FF industries (even more than now). If we persist against that kind of pressure, we'll also have the public policy moxie to push EVs harder.

I don't think you understand the shift happening in the car industry. It takes a while to develop new vehicles, but that's been happening for the last 4 years. Now, all of the big car companies have plug-ins of various sorts coming between 2010 and 2013. That provides a large resource which can be ramped up pretty much as needed.

we're certainly headed for our next leg down when the next big oil price shock arrives, which appears to be sometime between now and 2013 or so. Certainly by 2015.

That could have happened if OPEC had allowed oil prices to go back to $50, and if the world economy was growing more quickly. Doesn't Colin Campbell now predict that oil prices will stay below $100 for the indefinite future?

even before then, the world economy is queued up for contraction as this credit bubble continues to burst.

Can you point to any mainstream economists who agree with this? If not, what source would you consider most credible?

The Automatic Earth has just updated its primer.

1) Her site assumes oil is fundamental to economic growth, which IMO leads to a circular argument: collapse causes oil supply problem, oil supply problem causes collapse. While most of her discussions I've read are about finance, a careful reading can identify signs of this fundamental energy assumption. For instance, a discussion of regional differences betrayed an underlying reliance on assumptions about energy (areas with better domestic existing energy supplies will do much better). Other signs of her fundamentally unrealistic energy assumptions: the idea that suburban housing will crumble, and that oil could and will go to $500 or more.

2) Many of her predictions have already been falsified, e.g., that a post-2008 crash of oil prices & international shipping would be sustained. Similarly, she said the Dow would crash in 2009, and reach 1,000 in 2010. She predicted the Dow . The US stock market was up in the fourth quarter of 2009, and as of July 15, 2010 it was up for the year. She predicted sharp, dramatic price deflation - as of July 2010 all measures of prices showed growth.

3) She assumes deflation and post-bubble under-shoot can't be prevented. I referred her to Jim Hamilton's explanation of how the Fed could prevent deflation - she didn't really respond.

That's a choice.

Exactly. Which is why people are going to overwhelmingly continue to choose gasoline-powered vehicles over electric ones until electric vehicle technology can match the overall value of gasoline powered ones. None of the projections that I've reviewed show much of a dent by EVs before 2020. Here is another projection if you didn't like the previous one:
"But plug-ins, in both hybrid and pure electric varieties, will make up 1% or less of the global market - less than 1m units a year in 2015"

I don't think you understand the shift happening in the car industry.

I think you only look at the technology and fail to incorporate business thinking into your assessments.

Doesn't Colin Campbell now predict that oil prices will stay below $100 for the indefinite future?

Beats me. I go by the oil megaprojects database in Wikipedia for short term assessments.

Can you point to any mainstream economists who agree with this?

Well, I hesitate to go down this path because I think most economists are working with faulty models. However, if you absolutely need one, here is Krugman. There are some others, like Stiglitz and former IMF Chief Economist Simon Johnson. In any case, none of them include oil in their thinking, which is yet another major blind spot.

Her site assumes oil is fundamental to economic growth, which IMO leads to a circular argument: collapse causes oil supply problem, oil supply problem causes collapse

Yes, that's how the economy works. There are many feedback loops in our world, some positive, some negative.

Like I said, you should read their primer. You would learn a lot.

I have no idea about her other predictions being correct or not. She may have been incorrect with short-term predictions but the end game is clear: this credit bubble is bursting.

people are going to overwhelmingly continue to choose gasoline-powered vehicles over electric ones until electric vehicle technology can match the overall value of gasoline powered ones.

1st, EV's like the Volt and the Leaf already match the overall value of ICE vehicles, as do hybrids (the Leaf has a range limit, the Volt doesn't). Uptake of hybrids was and is slow because...they're new. New things take a while to be accepted, all else being equal (which they're not likely to stay, one way or another - you expect much higher oil prices, right?).

2nd, this isn't really a matter of individual choice, it's a matter of public policy. Most of the costs of ICE vehicles are external: security (think Iraq), financial (you suspect our oil dependency will cost us many trillions, right?), pollution (CO2, mercury, etc, etc), etc. When we decide to properly price fuel (via taxes, CAFE, cap and trade, or whatever) things will change.

None of the projections that I've reviewed show much of a dent by EVs before 2020.

Of course. They're based on BAU. You don't expect BAU, right? you don't believe EIA forecasts, right? Why would you expect conventional projections to be correct?

I think you only look at the technology and fail to incorporate business thinking into your assessments.

What do you mean?

I go by the oil megaprojects database in Wikipedia for short term assessments.

I took another look just now: I don't see an explicit production forecast. Perhaps more importantly, it doesn't tell us anything about changes in demand.

none of them include oil in their thinking, which is yet another major blind spot.

I'd argue that's the blind spot in what you're presenting. It assumes oil is irreplaceable, when it really isn't.

Yes, that's how the economy works.

No, it really isn't. There are no conventional economists arguing that we face TEOTWAWKI. The worst forecasts are for extended stagnation, rather like Japan.

Like I said, you should read their primer. You would learn a lot.

Ahem. I have read it. As I noted before, their arguments come down to their energy assumptions: that declining energy in general, and oil in particular, will cause extended economic decline. That's highly unrealistic.

I have no idea about her other predictions being correct or not.

You should. It tells us something about the accuracy of her model of the economy. When she says in November of 2008 that the economy might rally for a couple of months, but will then "continue" a dramatic decline, she's clearly not talking about the real world. Last year she predicted the Dow would be at 1,000 by the end of the year: what should we conclude when that prediction fails completely?

Now we're going over the same stuff (I already explained that I expect car sales to plummet, for instance, which is definitely not BAU) plus if EVs were an overall match to gasoline cars (same range, same cost) all these EV sales projections would be higher. That we can't agree on even simple relationships like this and that you aren't aware that mainstream economists don't include oil in their models means that we truly are very far apart in our understanding of how the world works.

Thanks for conversation.

Now we're going over the same stuff (I already explained that I expect car sales to plummet, for instance, which is definitely not BAU).

Oh, I knew that - I was just making a point: it doesn't make sense for you to rely on BAU forecasts, when you don't expect a BAU world.

if EVs were an overall match to gasoline cars (same range, same cost) all these EV sales projections would be higher.

Not really. Think about the Prius: it's better in most ways and doesn't have significant compromises, but it's still only a bit above 2% market share after more than 10 years. 10 years ago no one expected the Prius to do nearly this well. The fact is that in a BAU environment, change is slow.

you aren't aware that mainstream economists don't include oil in their models

Oh, I'm very aware of the difference between how mainstream economists handle oil, and the approach of people like yourself and Stoneleigh. I should think you'd remember our many conversations about Ayres, Hirsch, etc, etc. I just think the mainstream approach is much closer to reality and the TEOTWAWKI approach is highly unrealistic. I do think oil is important,and I think most economists would agree, more or less. I think Jim Hamilton's approach is very sensible, for instance: he is often cited on TOD, and he certainly doesn't think that PO equals TEOTWAWKI.

That we can't agree on even simple relationships like this...means that we truly are very far apart in our understanding of how the world works.

I think that if we were to have a really substantive, detailed conversation (rather than this kind of indirect discussion of "expert opinions"), we'd make progress. I have on many an occasion with others on TOD.

Oh, I knew that - I was just making a point: it doesn't make sense for you to rely on BAU forecasts, when you don't expect a BAU world.

Bad logic Nick. Andre is arguing that the BAU forecasts are unrealistic in a systematic manner, i.e. whatever BAU projections are for car sales, he expects them to be less than that, due to overall economic contraction. In a way, he is relying on BAU forecasts.

In a way, he is relying on BAU forecasts.

Precisely the problem: he's relying on forecasters who are embedded in BAU assumptions, like sufficient oil and static AGW-related public policy. In a world of scarce oil, increasing awareness of AGW, and accelerating investment in batteries and related tech, EVs will become dramatically more important. This will only be accelerated by an environment in which oil import problems are hurting the economy.

EVs will become dramatically more important.

Perhaps, but will they be affordable? If 'oil import problems are hurting the economy,' that means by definition that fewer people will be able to afford a new car of any type. That's the crux of the difference of opinion between you and Andre. Andre is anticipating a fast economic collapse scenario based on oil depletion and reduced credit (the two of which he sees as inextricable). He doesn't think that anyone except perhaps the very well-off will be buying new cars. I gather you think that however much you think 'oil import problems will hurt the economy', you don't think it will be too much to damage a market for tens of millions of cars over the next couple decades.

For my own opinion, I think you're both right. I don't know how badly the economy will be damaged, but I'm fairly confident that in twenty years there will be both a ton of EVs on the road (relative to now), and fewer total vehicles. I don't know if that means we (in the US) will have 10 million cars of which 5 million are EVs, or 50 million cars of which 25 million are EVs, but either way (or in many other ways) you could both end up being right.

Andre is anticipating a fast economic collapse scenario based on oil depletion and reduced credit (the two of which he sees as inextricable).

Yes. The thing is, that makes no sense. No matter how fast oil production declines, we'll still have enough to provide the physical energy needed to do all of the important things: produce food, transport freight, get to work, build and install wind turbines, manufacture EVs, etc, etc. We may have to carpool, and vacation closer to home. We won't lose anything important. I don't think Andre understands that.

Recessions can be caused by fear: fear of loss, fear of investing, fear of spending, etc (other causes include reactions to economic imbalances, like the recent housing bubble - that doesn't apply here). The argument for these forecasts of TEOTWAWKI that comes closest to plausibility is that PO will scare people so much that they, and the economy, will become paralyzed. That would be a case of a self-fulfilling prophecy. Fortunately, most people are much more sensible than that.

No matter how fast oil production declines, we'll still have enough...

Well, you don't mean that literally. ;-)

The argument for these forecasts of TEOTWAWKI that comes closest to plausibility is that PO will scare people so much that they, and the economy, will become paralyzed.

I think the argument for TEOTWAWKI that comes closest to plausibility is that post PO economic instability will lead to war. I do indeed hope that most people are more sensible than that. (I'm not encouraged by the trends in US politics.) Here's hoping that you are right.

you don't mean that literally.

No - I just mean that the most plausible forecasts don't come close to decline rates that would truly deprive us of the oil we'd need for basics. Heck, even the most pessimistic forecasts don't do that.

For instance, we could reduce oil consumption for personal transportation (which accounts for about 50% of overall oil consumption) by 50% in a few months, with aggressive carpooling. That's a net reduction of 25% overall, and we could do much more in a year or two of work.

We have 2x-3x as much oil as we need for the basics.

the argument for TEOTWAWKI that comes closest to plausibility is that post PO economic instability will lead to war.

Well, that requires really serious economic problems. And, again, I just don't see that as likely. We can't rule it out - I didn't think the US would be so stupid as to get into Iraq the way we did. But, I think we've learned a little bit from that...

Hi Nick,

EV's like the Volt and the Leaf already match the overall value of ICE vehicles, as do hybrids

Last summer, our old car crapped out and we needed to replace it. Although we use our bikes/trikes as much as possible, we still need a car in this insane car culture. However, we don't put many annual miles on a car. Also, we can't afford to spend a lot of money buying a car. I liked the idea of a car like the Volt, but the price was totally out the question (even if we could have waited for it to be available for sale).

We bought a new Chevy Cobalt that gets around 35 mpg for $10K after all the discounts and incentives ($19K list) plus a tax credit for the sales tax. Even at $10 per gal for gasoline, I could not justify a Prius - probably not even at $20 per gal - because we don't drive that many miles per year. So, I think the value proposition really depends on many owner dependant variables.

Ideally, I'd like to see Neighborhood Electric Vehicles (NEV) combined with Human Powered Vehicles (HPV) and really effective mass transit. What I'd prefer and what will actually happen may well exhibit a disparity.

I had to go into "the big city" today and return at rush hour. Only a person suffering massive delusion could think this is a transportation model that we should preserve. It is quite incredible how the human mind can be manipulated to believe that destructive modes of behavior are somehow admirable.

I agree completely: trains and bikes are a much more humane way to travel. where possible. I do my commuting by electric train, and drive only about 2,000 miles per year.

OTOH, it's important to remember that's separate from our energy problems: we have plently of cheap electricity to power conventional sedan (or even truck) EVs, and there's no reason to expect that to change.

Hi Nick,

we have plently of cheap electricity

I sincerely hope this proves to be true over the next decades. Regardless of liquid fuel shortages, I would have minimal personal problems if we have an affordable and reliable electricity supply - can't speak for many other people/companies that are more dependent on liquid fuel. But, lack of electricity would be a real game changer.

I worked in India (Bangalore) when electricity was fueling a dramatic change in their economy. Power interruptions occurred several times a day (nearly every day) so UPS devices were essential - electricity was a highly cherished commodity. It helped me appreciate what we take for granted.

I sincerely hope this proves to be true over the next decades.

The thing is, oil is the only fossil fuel that's in short supply any time in the foreseeable future. Almost everywhere, if you look at all hard you find enormous amounts of coal - the continental US, Alaska, Australia, the UK, all have enormous amounts of coal. China may be more limited, but I'd be a little surprised.

So, as far a electricity goes, our only real problem is CO2 emissions. For that, we have wind, which is cheap, scalable, high E-ROI, etc. We can keep using just a little gas and coal for any balancing of intermittency that's needed, or biomass, or synthetic hydrocarbons, if we want.

"If you look at all hard you find enormous amounts of coal - .... the UK"

I quite often hear that said, but can't understand it at all.

UK proven reserves are 400 million tonnes, annual coal consumption now (i.e. with coal contributing quite a low proportion of total energy) is around 60 million tonnes (mostly imported obviously) and even that is being very generous to the UK because the average CV of UK coal is well below that of the international coal that we are actually using.

Even if you triple our proven reserves by applying a hugely inflated price, and assume they could be produced at will, it's less than 2.5 years of UK's BAU primary energy demand.

Shockingly, our consented reserves are less than 45 million tonnes - less than a year's consumption at BAU.

Even if the UK had no other energy source, I doubt we could produce much more than 50 million tonnes per year and that only with massive government-led effort and investment, and then only for a decade or so.

It's true that Australia and Indonesia in particular still have vast reserves available for export, while China, Russia and the US have enough reserves to keep them sweet for a very long time, maybe 100 years or so, depending on demand growth, but Europe has no such option.

Mostly twaddle, as evidenced by the fact that the article was from 3 years ago and nothing at all has happened since. I'm afraid I don't have much time for grand plans announced by academics with no financial backing, they dine out on the initial publicity and lose nothing when it fails to deliver (that's always someone else’s fault, usually the taxpayer for failing to provide limitless high risk capital for free).

I'm afraid I was personally involved in negotiating the first ever coal import contract with the Port of Tyne, thereby literally "bringing coal to Newcastle". Psychologically, it was not a happy day for the city, but to be honest they were very glad for the business - the last deep mine in the region at Ellington near Lynmouth had closed earlier that year (2005 – the guy who used to buy almost all their coal is a friend of mine). There are now some very small open casts left in the North East that are permitted, we are talking mines that will produce 1-3 million tonnes in total here, miniscule by global standards.

There is still significant coal in the North East, some of which are proven reserves and included in my 400 million tonnes, some of which isn't but could indeed be extracted by gasification if prices reached unheard of levels (and I guess that would be included in my triple x proven calculation) and a lot of which will likely never be extracted no matter the price, due to EROI or land use change. All the best seams were mined long ago, and quite a lot of coal was actually exported from the region.

Use of the word "vast" in that article, is a bit of a giveaway, since no actual numbers at all were quoted. This line in particular made me smile: "the North-East was sitting on some of the best coal reserves in Europe". That's like saying you've got a degree from the best university in Central Africa or you're serving the best wine grown in Scotland.

Mostly twaddle, as evidenced by the fact that the article was from 3 years ago and nothing at all has happened since.

There's an enormous difference between what's competitive (and therefore actually likely to happen), and what's feasible. Industrial civilization could operate just fine on a price of $.40/kWh for energy, and that could support much higher prices for coal (especially with higher efficiencies),if needed.

As Yergin ("The Prize") tells it, UK coal was partially displaced by cheaper oil. It's my understanding that later cheap coal imports, combined with conflict with the coal miner's union (especially between a militant, strike-prone union and a conservative government), killed most of the rest of the domestic coal mining industry. Later, cheap domestic oil & gas kept domestic coal uncompetitive. Does that sound right?

There's an interesting discussion of historical coal reserve estimates here (I ignore the climate change POV):

I know much more about US coal than UK coal. It seems clear to me that the US has far more coal than most "Peak Coal Revisionists" recognize, especially in Illinois and Alaska. That makes me suspicious of what they say about other countries.

EVs will represent a minuscule portion of the vehicle fleet by 2020 even if the economy stays in its current limping along condition

Do they have to?

Suppose we decide to promote something like Trev.  The base model Trev might carry 4 kWh of Li-ion cells, enough for about 40 miles of range (the median daily commute is ~22 miles).  At $500/kWh, the battery is $2000; the complete vehicle might go for $14,000.  The market at that price would be huge.

Promoting Trev-like vehicles with licensing and insurance breaks (e.g. swap the plate off an ICEV at no extra cost) might push several million a year out the door.  Call it 5 million, starting in 2012.  That would give 40 million on the road by 2020, which is a substantial fraction of the total US LDV fleet.  If they were used at an average of 25 miles/day, they would account for about 1 billion miles/day or 360 billion miles/year.  That's about 12% of total US LDV vehicle-miles, and probably the most important 12% (we can cut pleasure trips).  Recharging at work could allow regular use for commutes up to 50-60 miles.

Power demand wouldn't be bad.  At 100 Wh/mile, a 25-mile/day route would take 2.5 kWh or about 100 W continuous.  40 million Trevs would take 4 GW; US wind generation is adding that much every couple of years.

Lithium demand wouldn't be that bad.  20 million kWh/yr of batteries would take perhaps 40,000 tons of lithium carbonate.  Simbol mining declares it can get 16,000 tons of Li2CO3 from just one geothermal resource in California.

More difficult applications can be electrified also, like Alt-E LLC's converted F-150.

Batteries for EV are optimized for weight and volume, not for durability and cost. Dividing the cost of a Li-ion battery by the total number of cycles it can bear will still give you a price higher than a standard production price from hydro or nukes ! It means that even if intermittent power was free, it would still not be worth using your car battery for load balancing ! The reason EV still makes sense despite the high cost of their batteries is that we are ready to pay much more for mobile transportation energy than for static heat or electricity. The petrol tax that European pay (around 50 cents p.l.) is roughly equivalent to a price of carbon of 250 EUR per ton ! (Incidentally, it gives an order of magnitude of the prices necessary to completely kill the CO2 habit through pure pricing mechanism...).

even if intermittent power was free, it would still not be worth using your car battery for load balancing !

Dynamic charging is free. IOW, it doesn't cost any more to charge at 2 am, when wind power is cheap, than it does to charge at 3 pm, when electricity costs are at their peak.

Now, V2G would mean using charge/discharge cycles, but those would still be cost-effective for unusual events where costs spiked and additional capacity was needed. A li-ion that costs $350 per kWh and gives 5,000 cycles gives a cost per discharge of 7 cents. That's pretty cheap for instant backup power.

I have read some discussion of ammonia synthesis as a "dispatchable" load. In a post carbon world ammonia will need to be synthesized in some manner. Does anyone have any data on the costs of operating ammonia plants in an intermittent manner?

I'm confused by the "Diurnal variations of power generation, France" graph. From the graph above it, it appears France exports electricity at all times of day, with maximum exports between 3 and 5am (which makes sense), but then after you "subtract the minimum" this peak export flow shifts to the middle of the day. I am guessing a sign error, but please clarify if I am missing something. Interesting post, thanks.

Yes it's a sign issue. Exports are a negative contribution to cross-border flows:

Domestic consumption = domestic production + imports - exports

Net cross-border flows are (imports - exports), hence, if there are net exports, the minimum of cross-border flows is negative, and is at the maximum of net export.

The major user of electricity, and peak component, is air conditioning. Wind does not correlate very well, solar does (i.e., more sun, more A/C). One solution is to hook everyone's HVAC system to a communications network and throttle the A/C down/up as power is available. So when the wind is blowing and sun is shining, your A/C unit will be pulling the temp. down to 70, effectively storing energy in cooled air. During a drop in wind/sun, it can idle and go up to 75. The incremental cost of this equipment would be negligible for the vast majority of Americans- a couple bucks for the wireless chip and some additional logic in your thermostat. The internet can provide the connection.

Good news is there are solutions that will keep us from reverting to pre-oil days. Bad news is by jerry-rigging price signals, the better solutions may be overtaken by inferior ones. Imagine if Jimmy Carter would have had a major initiative to put an IBM 360 mainframe and terminals in everyone's homes- we'd have spent a lot of money on an inferior solution.

(By putting air conditioning on top, I am guessing you are referencing South US?)

There are more direct ways to apply solar power for air conditioning: the Wikipedia article lists several methods. There are other heat exchange systems too, and above all, better building construction.

The correct term is "jerrybuilt" which, BTW, doesn't have the same meaning as the term "jury-rigged." Jerry-rigged is therefor a "mixed metaphor."

Storing "cold" locally for later use locally, when excess electricity is available isn't a bad idea for peak shaving. Some "cold" is actually stored as ice cubes, but this seems to be little colder than desireable to be very "thermodynamically efficient". Maybe some other phase-change material would be better than ice, but the engineering of systems using these materials isn't as straightforward as you would think, if memory serves me correctly.


70!!!!!!!!!!!! That is firkin freeezing! 80 is more like it with 75 being COLD :) These are based on actual temperatures here. Also morning there is not much use for A/C but the demand is well into the night. If a PV system could cool down a large tank of water to supply the cool for the rest of the day it would not be a bad idea though. Sometimes we need to get out of the thinking box that cheap electricity has put us in and think in terms of storing energy. Store cold for your A/C or fridge, store heat for heating. Much better efficiency to the double or triple conversion through electricity and batteries. Just de-humidifying the air, here, would make living very much more comfortable with much less need for out and out cooling.


Actually, I like thinking in terms of lots more cheap energy, and how to make the best use of it.

As opposed to having windfarmers turn their blades sideways since there's nowhere to use the power they're generating. And ending up with the same situation here that they currently have in China- all the wind companies losing money.

70!!!!!!!!!!!! That is firkin freeezing! 80 is more like it with 75 being COLD :)

I think it varies all over the map. A lot of commercial outfits keep the temps really low on hot days to lure in hot shoppers, and employees may be seen wearing sweaters.... In terms of dwellings, I know some who easily tolerate 82, and some would wouldn't dream of going about 70.

Sitting here it is about 83/83 (F/%) and even in a pair of shorts it is uncomfortable. Once the storm breaks it will be much more pleasant. Now if that was 83/73 or 83/63 it would be far more comfortable. You don't need freezing temperatures to make life more comfortable. Humidity control can be achieved far more economically. As for attracting customers I knew a sales manager who kept the room freezing cold so that potential customers would just sign up to get out of the cold. Guess what? The potentials did get out of the cold as fast as they could, didn't stick around long enough to buy.

Oh, and if there is lots more cheap wind energy than can be used then how much coil/oil/gas is really needed. The problem is not one of excess wind but one of excess FF.


If you keep signing your comments that way, you're going to get a nickname like Naom Chymsko. ;-)

Err, did you mean Noam Chomsky? ;) I chose my nym to make it clear I was not in the oil business. There were a lot posting telling exactly what BP should do and trying to make out only they had the special knowledge to fix the well, they would turn out to have absolutely nothing to do with the oil industry. So I decided to make it clear and NOAM is a nice short abbreviation for people insted of notanoilman though Rockman tends to use not when he helps my learning experience. When I first came here I was surprised to see people talking about the things I was wondering why people were not talking about so I thought I would stay.


Hi Luis, thank you for that post. The role of mid merit plant (what you call intermediate load) is of course central in grid dispatch and is becoming ever more so with the rate of penetration of wind and other intermittent renewables. No grid can possibly hope to be stable without it, bearing in mind that mid-merit plant also provides most of frequency response and spinning reserve which is essential to match loads on a second to second basis - again ever more challenging with greater wind penetration.

In the UK, mid-merit has tended to be provided by mainly pulverised coal plant and older gas CCGT, and it has required most mid merit plant to go much further than simply ramping up and down. With a couple of exceptions all UK coal plants run 2-shift regimes (i.e. turn on and off every day) for 6 - 9 months of the year. In fact this year several coal plants 2-shifted right through winter although that has partly to do with running restrictions under the Large Combustion Plant Directive.

In a post fossil world, large circulating fluidised bed (CFB) boilers running on biomass are very well suited to perform this role. Due to the large inert mass of sand in these boilers they are very well suited to retaining heat during a 6-12 hour shutdown and re-starting quite easily. An appropriate fuel for these boilers is wood chip as it is low in chlorine, ash and heavy metals which make certain types of biomass challenging to combust.

You lightly dismiss "biofuels" as having a low EROI and being subject to land constraints, and these are both genuine worries with regards to certain types of biofuel, especially those that rely on primary agricultural outputs such as cereals or oilseeds. However they simply do not stack up for solid wood biomass, let me take the two points individually:


The largest scale biomass boilers being contemplated are about 300MW net capacity, at this level these plants can achieve 40-42% net conversion efficiency, depending on the cooling that's available. There is not a lot of difference between a biomass plant and a hard coal plant, except that no boiler manufacturer today will go supercritical on biomass because it is a step too far to deal with alkali ash melting and corrosion at those temperatures at the moment. However, once they have more experience on with big boilers they will get there, probably by using additives to control the chemistry. A port based 300MW biomass CFB burning imported wood chip from North America (where there is massive surplus potential) has an EROI which is roughly similar to our current coal plants (burning Russian coal at an average net efficiency of 36%) and very slightly lower than an average aged CCGT burning Qatari LNG Siberian piped gas. As a rough rule of thumb the energy required to ship the wood chip (using panamax class wood chip carriers) is about 1.2% of the biomass energy in the hold assuming the ship returns empty. The energy consumed in harvesting, chipping, transport to port, loading and discharge is about the same again.

Land constraints:

The underutilised resource from existing European and Eastern North American working (i.e. not protected) forests that could be sustainably harvested or collected and supplied to a European port is around 250m tonnes (depending on who is doing the estimating, we have seen much larger figures used, but this one is much closer to what is realistically practicable). That would comfortably supply about 100 of the large 300MW boilers that I mentioned earlier or about 30GW if they were running baseload. However if they were 2-shifting with say a 40% loadfactor (typical of mid-merit coal plant in the UK today) that would be about 60 GW of plant. A very respectable % of Europe's mid merit needs.

However the real resource potential is South America and Africa where huge areas of degraded land (typically used for a short period for agriculture and then abandoned) exist which would be used for sustainable forest plantations. Brazil alone estimates about 110 m Ha of abandoned or unutilised agricultural land that would be suitable for forest plantations (i.e. there would be enough rainfall), energy optimised plantations (as opposed to the pulp fibre optimised variety where thee genetics are pushed to produce as much cellulose as possible) can easily achieve 60 tonnes per net planted Ha today (and likely much more in the future). Assuming 50% net to gross planted area on 110m Ha of land that would equal enough fuel for almost a terrawatt of 40% LF mid-merit biomass plant. That alone would be far more than enough to cover the mid-merit needs of the whole of Europe and South America, and doesn't take into account other vast areas of potential in South America and Africa.

I stress that this is all land that would literally benefit from having trees grown, even though it may be a mono-culture and therefore not environmentally perfect it would still prevent soil erosion and economic depredation on a massive scale (which usually leads to overexploitation of soils and other natural resources such as high biodiversity forests) and preserve the value of the land for the long term. Most of this land is not suitable for arable crops, or at least would take massive energy-intensive inputs to be so. Soil preparations are mainly potassium and phosphorous and other micronutrients, i.e. not heavy nitrogen. They can mainly be recycled from biomass ash. By the way, North America could already produce enough from existing sustainable forestry to meet its mid-merit needs.

Biomass is not "the answer" to future power needs. Rather it has a role to play role along side baseload nuclear, variable wind and solar, and flexible hydro. Although due to subsidy regimes, the first generation of large dedicated biomass plant will be baseload (the grid has all the mid-merit it needs for now with coal and gas and not enough low carbon generation) the future of biomass is to fill the mid-merit niche, while providing traditional “thermal” spinning reserve and frequency response. It is a vital role if we harbour any hope at all that power grids can be kept stable at acceptable prices in the future, and does not deserve to be so lightly dismissed.

Boimass as fuel will never likely overcome the problem of extremely low efficiency of photosynthesis. eg. we're better off rather than any presently available biomass system, to simply cover a small portion of the proposed growing area (<1% efficient insolation-to-biomass) with solar-thermal collection (>15% efficient insolation-to-electricity).

Dead case. May have a few niches, but no scale.

That's relevant for problems of scale, but not for cost. Biomass for burning in a thermal plant can be extremely cost effective. Heck, the US gets about 1% of its electricity from wood right now. That's more than wind until very recently.

Lengould, I don't disagree that there is a place (probably) for solar thermal, but for the reasons I have mentioned above there is also a place for large scale biomass. They are not really in competition. The amount of land required for solar thermal is insignificant compared to biomass and the most appropriate land areas don't overlap much given that the best place for large scale solar is desert (pure desert is a step too far for even the most drought tolerant trees). Conversely the amount of investment and energy required per Ha is massively more for solar thermal.

However solar thermal can not by itself run a grid, I know there are arguments that heat can be stored overnight to make solar thermal closer to baseload, but realistically such a method of storing energy is way down the merit order in terms of cost effectiveness (certainly way below biomass) and therefore not a viable argument.

To rehearse the points again:

1. Solar is a high capex venture and that means that for solar to work economically it has to produce power whenever it can, in other words it fits into the intermittent source category (or at very best baseload) that Luis described. As Luis pointed out, more flexible plant is also required on the system to match supply and demand, biomass has a much lower upfront capex cost but higher unit costs which makes it well suited to the mid-merit role.

2. Solar can only perform at certain times of day, a power grid can not survive on this alone, it needs reliable back up sources of power, such as can be provided by biomass.

3. The source of solar is limited by location, i.e. the power has to be generated where the land is. Biomass can be transported so that the power generation is much closer to the source of power demand. This matters, transmission constraints and power losses are a big consideration in operating a grid.

4. Ancillary Services. Running a grid requires sources of frequency control, spinning reserve and other hot reserve. This is plant that is available to respond to a sudden ramp ups in demand (such as a tea break in world cup football game) or drop in supply (such as plant failure or e.g. a huge dark storm cloud moving over a large solar installation). Frequency response is plant that is able to automatically detect the drop in frequency from the system and respond by increasing generation quicker than any human intervention could manage (less than a second). Spinning reserve is part loaded plant that can also respond but on a slight longer timescale (several seconds) and hot reserve is warmed up plant not actually generating and usually takes several minutes to respond. All of this requires reliable thermal plant running at part load. Unreliable solar and wind are not suitable for this, and high capex makes the idea of part loading this plant economically unfeasible. In fact their presence on the grid means that even greater levels of frequency response and other reserves are necessary.

Simplistic arguments that a grid can be run on pure solar and wind are nonsense (at least without insane, civilisation-destroying levels of cost), and are frankly depressing. There is a place for intermittent renewables which can contribute to net energy creation, but there is no way a grid can be kept stable without secure and reliable sources of thermal plant for which, in the long run, only biomass and nuclear qualify, and of the two, biomass is the best suited to mid-merit due to capex versus unit cost considerations.

Simplistic arguments that a grid can be run on pure solar and wind are nonsense

The only people I've seen making such arguments were, in fact, opponents of solar and wind (or, as in the case of a recent TOD article, making unrealistic arguments for TEOTWAWKI). It appears to be a strawman.

Fair enough, I didn't mean to cast aspersions. But Lengould's comment did seem to suggest that, at least to me anyway.

you say:
3. The source of solar is limited by location, i.e. the power has to be generated where the land is.

Disagree, please read:

Europe Eyes Africa for Solar Power
Solar-thermal plants throughout the desert would send power across the Mediterranean Sea

Trans-Mediterranean Renewable Energy Cooperation

Will North Africa come to Europe’s rescue in 2020?

Desert Power: The Economics of Solar Thermal Electricity for Europe, North Africa, and the Middle East

Mauisurfer, that's right but you have misunderstood my point. Of course you can build a plant in any desert you wish and then build a great long transmission line to ship the power to the point of demand. With solar thermal you HAVE to do that, since there are very few large cities in deserts outside of the US.

But my point is about actually balancing the system on every piece of transmission equipment (wires and transformers), it may be a slightly esoteric point to those who haven’t worked with power grids and I don’t want to lecture any more than I already have, but it does take work and require flexible responsive plant to match up power flows over transmission lines without exceeding fault levels. The further the distance between supply and demand, the more true this is and the more vulnerable to faults the system becomes, and therefore the more flexible, responsive plant is required closer to the demand. For example, in the UK (a small and crowed country with one of the most long established grids in the world) the system operator already has to constrain plant on and off to the tune of several GW on a regular basis because of transmission constraints over north-south power flows. To do this he needs responsive thermal plant.

If trees haven't retaken the land as fast asit was abandoned by farmers it is not likely they will grow very well, if at all.

It might be possible at great expense to establish something approaching woodlands or forest where it doesn't establish it self fast, but the odds aren't good.

There will be no easy lunch to be had here.

Oldfarmermac, decades of plantation development history and uninterrupted yield improvements in Brazil, Uruguay, Chile, South Africa, Australia, the US South East and other regions globally would completely disagree with you on that one.

You are right that once the land is abandoned, scrub will regrow, it won't remain literally barren. But the scrub is not of much value to anyone, it is scarcely more biodiverse than the plantations which replace it and the biomass growth rates are poor. The plantations achieve dramatically higher levels of growth by two main ways:

1. Ground preparation. The most usual reason the farm land was abandoned in the first place was poor soil that did not support grazing for long or many crop rotations. The science of ground preparation is impressive, it usually focuses on potassium and phosphorous but other nutrients also come into play.

2. Genetics. The trees that are planted have been bred over decades to match the right conditions and maximise yield. In Brazil, many of the plantations have been created in order to produce charcoal for pig iron production and are therefore optimised for energy yield. Others are optimised for cellulose yield for paper production.

There are other important factors such as protection against ants for seedlings which are more location specific.

I don't deny that these are mono-cultures and therefore not environmentally perfect, although they do support an impressive array of wildlife, but they are still much much better than doing nothing, a case of not letting best be the enemy of good.

Hoover, I don't really disagree with what you have to say, but rather about the scope of the practicality and difficulties involved.It's more a matter of what each of us means by given words than disagreement.

Land that is suitable for trees will certainly very quickly be invaded by trees, as any forester or farmer will tell you, as soon as somebody STOPS doing whatever kept them off the ground in question.

Scrub, as you call it , is temporary, and is soon replaced with forest without intervention being necessary-if the land and climate are suited to forest.

Certainly tree plantations can be established-I own a (rather small) one myself.

I live in an area with many commercial tree plantations and have traveled quite a bit in the southeastern US,and seen lots of them. Excepting a few located on very steep land prone to erosion , virtually every one of them sits on land suitable for pasture , hay, or crops, given good soil management practices.

This former farm land was not "abandoned" so much due to poor productivity as fallowed permanently due to EXCESS productivity;it simply wasn't needed anymore for crops, as the markets were glutted, and farmers were going broke raising unsalable crops.

But establishing such commercial tree crops is labor intensive, and takes a comsiderable bit of money too, considering that any harvest is many years in the future.

I probably failed to make my point clear;what I wished to emphasize is that land that was not formerly occupied by forest, or land that is desertified, cannot be expected to support highly productive woodlands,either natural or cultivated, without making a considerable capital investment in establishing the monoculture tree crop.And even then, productivity will be strictly limited by prevailing natural conditions.

Tree growth is likely to be severely limited on abandoned farm land in most places for the same reasons the land was abandoned, the biggest one being lack of rainfall.

Of course some land that is abandoned by farmers is ideally suited to trees, and quite fertile enough for them to grow well.In our particular case, we planted white pine on land that was highly productive but uneconomical to maintain as pasture or hay land-its former use- given current day wage / market conditions.

I tend to get rather annoyed by reading about schemes to do things on the land that any body with day to day experience and a little professional training recognizes as unworkable..except at great cost with possibly disastrous long term consequences.

We aren't ever going to harvest huge quantities of switch grass off of dry marginal lands, especially without applying lots of fertilizer, because such lands don't, can't, produce large quantities of biomass quickly;and ANY plant removes nutrients when it is harvested and hauled away.The yields of these "new crops" will decline just as fast once they are harvested regularly as yields of conventional crops such as corn-unless the nutrients are replaced; and unless extensive and expensive measures are taken to protect the desired species from invaders.

If you selectively harvest a dominant species, such as a grass,others will quickly take advantage of the new conditions and invade and compete for water, light, and nutrients;this doesn't matter too much for trees though, as a tree monoculture once established shades everything else out effectively for the life of the tree crop..

I see articles in respectable science magazines and newspapers about growing our food in high rise buildings;certainly this COULD be done, but actually doing it would be utterly absurd, considering the costs involved.

Of course the egghead professors and architects and utopian planners who come up with such schemes don't have much to say about the costs.

Even so called marginal land is in short supply in most parts of the world.

That land needs to be kept in use producing whatever food it will for the people who live on it.

But yes, if one has the resources, it can be, as you correctly point out, be farmed with trees as the crop, using the same basic techniques used by conventional farmers;the only real difference is the time scale.

But that marginal land will no more produce large crops of trees, compared to good land, than it will corn or wheat or any other crop.

Biofuel crops are an enormous threat to the food supply.Because the labor requirement for tree farming is minimal except during the establishment phase,and the harvest phase decades later, poor people get displaced, and thier food supplies, already critically short in too many places, get even shorter.

When the tree farmer, especially the corporate tree farmer, gets his nose in the tent, the definition of marginal land changes as the market changes in his favor.

My ancestors were largely forced out of Ireland not because the land was unproductive as they were farming it, but because thier landlords could make more money on sheep and cattle.The potato blight was merely the final straw.

We can talk in theory for a long time but I would invite you to visit the same places in Brazil that I have visited.

These lands are scrub, and are not being farmed in any meaningful way. Soil quality and seasonality of rainfall (which trees with the right genetics can cope with) are the main reasons behind that. The people there can not afford the N (and KP) fertilisers, or seasonal irrigation, that would be required to farm food crops productively and they could never generate the net revenues that would be required to keep buying that fertiliser every year, because the land would not be productive enough.

The land possibly would return to genuine forest if left alone for a century or so but the desperately poor people that live there now tend to chop down any tree as soon as it reaches a certain size. You can't blame them. It is the law in Brazil that plantation owners are required to keep some areas of natural vegetation (20 - 80% of total depending on the state) and to protect it so that its natural utility can increase over time.

When a plantation comes it does create several thousand local jobs. Housing, education and health clinics come with it as well as roads and improved infrastructure. You can go and see these things for yourself, I invite you.

We can speculate what would have happened to the millions of acres in the South East US if they were not now commercial pine tree plantations, but it can never be proved either way. Many were former cotton or tobacco plantations that were no longer profitable due to global trends or where the land was increasingly less productive (tobacco in particular is brutal to the soil). Much it was unproductive until US government subsidies for forestry establishment came along in order to help the people and provide a future for the land. Most of the plantations in the US are optimised for saw timber, pulpwood and biomass are effectively by-products. It is probably the most sustainable and environmentally friendly industry in the US today creating millions of tonnes of building materials, that would otherwise come from steel or concrete, each year by literally sucking carbon out of the air, whilst also providing habitat far far more valuable than intensly farmed food crops.

Since we can not prove it either way, I can respect your view on US South East, but also disagree with it. My feeling is that it would have gone the same way as the abandoned pastures in Brazil. Natural scrub would have come and some people would have stayed there and eked out a desperately poor lifestyle that would have continued to degrade the land.

As for the costs of establishing plantations, these are now very well known and fully factored into my calculations. It is not theoretical, we are working with companies who have already established millions of hectares of tree plantations all over the world and real live (and expensive) power projects in the UK that will take the product and create a source of energy that can be sustained as long as there is sun and rain.

Hi Hoover, great comment, much appreciated - thank you for your considered response.

Is it possible to recycle the phosphorus from the ashes of wood burning power stations? (something to put in the ships on the return trip) You can only regrow forests on damaged land if phosphorus reserves are high enough. In the counties you mentioned this is propbably still true. But it does not seem to be a sustainable solution to me without the return of the phosphorus to the forested plantaions. Have I missed something?

Hi Tirwin, thanks for that good question. We have a programme looking at recycling of ash back to plantations. P is one of the most important elements of that along with K.

P depletion is very often a problem with the soils we are dealing with and often needs treatment prior to the first plant.

However most of the P recycling is done in situ, since it is most concentrated in the leaves and roots, which are left behind after harvest. Good practice Euc plantations on fragile soils will also debark in field leaving the bark behind which has the next highest P (and K) concentrations in the tree. The stem wood is very low, but there is still value in recycling the ash because it is so concentrated. In general P requirements are much less on second and successive rotations.


If I read your comment correctly,you are saying that wood can be harvested, processed, and transported to dockside for a little over one percent of its energy content.

In which universe? ;)I don't doubt your sincerity, but over the years I have found such figures to be highly suspect, and often off by a factor or two or three or even four or more.

Of course even if the energy costs total ten percent, the scheme might still fly, given the very high value of electricity when the chips are down, so to speak.What's another three of four cents per kilowatt hour when you are already paying thirty or forty cents-a likely price in the future where it isn't already a reality?

I am a rolling stone ag guy and have spent days watching my own timber being harvested with the most modern high efficiency machinery available, from flat land requiring no road construction or remediation ,other than leaving seed trees.My back of the envelope estimate is that diesel fuel alone would consume well over your one percent when you allow for a long truck haul, and a heck of a lot of the hauls , if such a scheme is ever implemented,will be long ones.

But it won't be, barring miracles of political skullduggery;the enviromental costs would be staggering, and won't be tolerated-until the wood is needed locally to keep local lights on.

Beyond that,what are you proposing to use to pay for all that wood?

Personally I forsee a rather strong market for it domestically, if the economy remains on its feet;and my guess is that protectionism will make a big comeback as well.

Americans can be more easily fed by paying higher costs for locally manufactured goods that they can by importing the goods and supporting them on welfare-not to mention the fact that the rest of the world will sooner or later realize that our checks are no good.

You'll have to forgive this, it's very rough, but I wanted to reply and the analyst in my company who actually does all these calcs is away, so this is very back of the envelope but should be the right order of things anyway:

Firstly, I assume a procurement radius of 50 miles to my stockyard. That gives an average haul of 35 miles (forgetting the bias towards nearer tracts).

We chip in-woods (for max biomass utilisation), the chip van gets about 9 miles to the gallon of diesel. It comes back empty, and so travels 70 miles and uses 7.8 gallons of diesel. A gallon of diesel has about 0.15 GJ of energy (sorry, I use GJ, it's about 139k BTU), so the chip van uses 1.4 GJ of diesel energy in its journey.

The van loads 27 short tons of chips of 50% moisture (it's slash pine) which is equal to 24.5 metric tonnes at a NET CV of 8.43 GJ per metric tonne. The loaded energy is about 205 GJ, so the energy to haul the chips is about 0.56% of the energy hauled.

After that it gets more complicated, and I can’t list all the calcs but I hope you can see the order of mag is about right:

Cut/skid/load = 0.21% of energy loaded
Chipping = 0.14% of energy loaded
Receive/stock/reload to barge = 0.03%
Phytosanitation = 0.09%
Barge 120 miles to ocean port = 0.29% (barge does not return empty)
Unload barge/stock/reload ocean vessel = 0.03%

Total = about 1.35% energy up to FOB on the ocean vessel

As for protectionism, it could happen yes, hopefully it won't. As human beings we can choose whether to voluntarily impoverish each other in that way or not, I remain inspired by the hope that we will not, but I also remain working very hard to make sure we have alternative supply routes - there is nothing special about the US, wood grows well everywhere, it's just that the industry is well developed and there are a few large corporate land owners who will sign up to long term supply contracts.

One important utility/grid management tool that has been neglected is overbuilding.

The current grid is overbuilt: the US has about 1,100 GW of capacity, for an average demand of about 450GW.

Obviously, it's cheaper to overbuild generation that has a low capital cost, like natural gas, but some modest overbuilding can make sense. France has done it with nuclear, and a grid with no fossil fuels would be likely to do the same with wind and solar. Unlike nuclear or solar, wind power can be turned down quite easily, so it could load follow. Just a small % of overbuilding would provide a dispatchable margin using wind power for most of the time.

On the other hand, the surplus power from an overbuilt wind and solar resource could go into various forms of storage: pumped storage, CAES (air), EV batteries, ammonia, hydrogen, synthetic liquid fuels: the list is pretty long.

The combination of an over-built wind resource and Demand Side Management using electric vehicles might eliminate the need for most other forms of generation for load following. The primary need for other forms of generation would be for supply diversity and seasonal variation. The combination of stored power (discussed in the previous paragraph) with biomass would provide "filler" for the occasional seasonal problem.

You can't say a grid is overbuilt unless its capacity is substantially greater than its peak demand--average demand is not the relevant comparison.

Yes, I'm using the word slightly unconventionally. Perhaps another would be better - any suggestions?

I think the problem is not your terminology, but that the capacity present must be able to meet maximum (peak) demand, not just the average; and that at any time (i.e. you have to provide for power plants down for maintenance or accidents).

No question about it: the US grid is not substantially over-built from a traditional perspective. I didn't mean to say it was.

Nevertheless, there is much more capacity than is needed to handle the average load. There are costs associated with that, and we accept those costs as reasonable to deal with demand variation. I'm suggesting that a similar strategy to deal with the production variation associated with wind and solar (aka intermittency) is also perfectly reasonable. In moderation, and in combination with other strategies, it is likely to be very cost-effective.

My use of the term "Overbuilding", even with clarification, appears to be confusing - any suggestions for one that is better?

So your point is that if, with demand management, the variations of the demand curve could be reduced, we could dump a lot of intermediate and peak load reserve capacity operated at low capacity factors?

Yes, I think we could greatly reduce the need for Fossil Fuel intermediate and peak load reserve capacity. We might still need some biomass, or legacy Fossil Fuel plants for seasonal backup, or for the relatively rare multi-day lull in wind/solar.

I was primarily talking in this comment about the value of "over-built" wind for handling load-following, but certainly demand management is also very important.

If we were to build 15% more wind capacity than the average needed, we probably could use wind for the diurnal load-following services now provided by peak generators roughly 75% of the time. Demand management could provide most of the remaining 25%. Exactly how far we go towards eliminating diurnal peak capacity entirely would depend on the exact costs, and optimization of the overall system.

My use of the term "Overbuilding", even with clarification, appears to be confusing - any suggestions for one that is better?

Maybe you just need to be more specific: "Overbuilding of intermittent (renewable) baseload capacity." Where we understand renewables as "baseload" in the sense described by Luis in the keypost.

One issue I wonder about is how cross border sales of electricity will hold up if some of the countries are having financial problems. Suppose, for example, Greece is the country wanting to buy electricity. Suppose too, that it hasn't been able to pay for the previous two months of electricity purchased. Who will want to sell it to them?

I picked Greece because of its reported financial problems, but there is a whole list of countries with problems. I can imagine a situation where several countries are not deemed credit worthy. How do you do deal with that situation? If Country A wants to sell to Country C, but to do so, the wires need to go through Country B which is not creditworthy, is this a problem? This starts sounding a little like Russia's natural gas problems, trying to export through the Ukraine.

Hi Gail, most countries don't buy power from each other, at least not directly, it is down to the power companies. Of course a lot do have state owned electric utilities with various levels of state interference (almost total in the case of Ukraine and Russia, less in France and practically none in Denmark and Sweden), but even the state owned companies have their own credit ratings which are seperate and different from their government's. Trading arrangements are usually such that all but the very strongest companies have to post credit before the power flows, usually a letter of credit from a bank (haha!), and settlement takes place either in the next week or next month.

The question is, would energy companies become insolvent just because their government and financial industry does? If there is a single currency, IMO that is not a high risk. Meanwhile, a recession might also reduce electricity import needs.

The example of Ukraine and gas is not well applicable here. As I understand it, the root of the problem is that originally, both Ukraine's transition tariffs and payments for own consumption were fictional: they were more or less meant to balance each other, and then various oligarchs tried to still make some money off the situation. No such situation exists in electricity transfer countries in the EU (though beyond the EU, I don't know Belarus's situation regarding electricity from Lithuania).

In Detroit Edison's service area, and presumably elsewhere, it is possible to hook up air conditioning to a second priority service. The cost per kilowatt-hour is less. The trade-off is, when the grid is heavily loaded, the electric company can shut off service to your air conditioner for 20 minutes every hour. It's a reasonable method of limiting their peak load.

In Russia, which I visited for three weeks a decade ago, I understand there is a hierarchy of priority customers. Presumably, when their grid is stressed, the lowest priority can be cut off before the whole grid collapses. How well it works, I don't know.

I believe we can expect to see more similar expedients in America and elsewhere, because it is obvious the capacity of the grid and of the plants that produce electricity for it is not being upgraded in the early part of the long emergency. Upgrades are even less likely as energy becomes even more expensive. Eventually, we won't expect continuous service from the grid, but will be compelled to adapt to intermittent service.

Or maybe we are already there, without quite admitting it openly. Last weekend's experience, of a severe brownout (45-60 volts on 120-volt service) that lasted for two days after a thunderstorm, demonstrated several things to me: the importance of having a small generator to run a refrigerator or furnace (the really critical items in a house); the functionality of incandescent bulbs when the low voltage would damage most low energy fluorescents; and the robustness of switching power supplies, which allowed us to watch tv while we were sweating out the brownout.

Perhaps a campfire series on practical ways to adapt to a failing grid would be useful.

In Detroit Edison's service area, and presumably elsewhere, it is possible to hook up air conditioning to a second priority service.

You're talking about Demand Management. It's very effective, and nothing like brownouts.

the capacity of the grid and of the plants that produce electricity for it is not being upgraded in the early part of the long emergency.

That's not realistic. Utilities continue to invest in the grid and generating plants.

In Detroit Edison's service area, and presumably elsewhere, it is possible to hook up air conditioning to a second priority service. The cost per kilowatt-hour is less. The trade-off is, when the grid is heavily loaded, the electric company can shut off service to your air conditioner for 20 minutes every hour.

That's the service I've got, and I've experienced it elsewhere.

Last weekend's experience, of a severe brownout (45-60 volts on 120-volt service)

You probably had a dead phase on a delta-connected transformer upstream of you.  The two sides connected to the dead phase will get half voltage.

the functionality of incandescent bulbs when the low voltage would damage most low energy fluorescents

I did a search and didn't find anything about this, at least for cold-cathode CFs.  I would expect them to fail to start if the voltage was too low, and otherwise to work okay.

I suggest that Europe's feed-in laws be repealed and the money spent on storage technology. That is a premium price could be paid for intermittent energy that has been stored somehow. There would be no additional tariff for real-time electricity production nor would there be obligatory purchase. However a CO2 cap would still penalise coal and gas but that cap should be tougher than the current EU version.

In other words the favoured sources of electricity production would be both low carbon and dispatchable. The present system is too soft on high carbon and too generous towards intermittent sources.

the money spent on storage technology.

Storage is very expensive. It makes much more sense to rely on geographical dispersion; careful site selection; demand side management; moderate over-building; diversity of generating types, etc.

These things aren't simplistic solutions, but they're much more effective, and much cheaper.

While I don't see intermittency as a problem as grave as you see it (also see Nick's reply), I also note that you ignore the investment stability and development angles of feed-in laws. Feed-in laws enable the industry to make new technologies cheaper. In fact, it would make sense to expand them to storage technologies, too.

Margaret Thatcher used to say that socialism doesn't work because one always runs out of "other people's money". Feed-in tariffs are just that : other people's money.
In some countries like Spain, the money well is starting to weaken : the government is balking more and more at the prospect of paying several billions euros per year to compensate for the facts that wind and solar are not cost competitive. In Germany, the price tag is even bigger (less sun than in Spain, and less wind than in Denmark).
A technology is ready for the market when people are ready to buy it without being coerced. Before that, it has to stay in the lab. I can understand that a pilot plant or a FOAK installation is supported by public funding, because at this stage, it is still applied R&D. But after that, it has to fly on its own economically. The 100+ billion that has been expensed with the mass deployment of renewables would have been much better spent paying researchers to do basic and applied science than feeding politically connected developers, bankers and industrialists.

Sigh. No, governments aren't paying feed-in tariffs: those aren't financed by tax money, but spread the costs to all consumers.

If you want to end over-pricing, you should also ban profits: for, on markets, you don't sell for the production price (that's a Marxist concept), but for the price set by the balance of supply and demand. On the electricity market, the price for all is the price of the most expensive producer needed to meet demand. Thus it happens that feed-in laws can reduce electricity prices.

The reasons governments change policy have a lot of reasons, including ideology and special interests. At any rate, the recent photovoltaic feed-in rate cuts starting in Spain have to do with just the kind of success I mentioned: stronger than expected (degression rate exceeding) price cuts, which made photovoltaic more profitable.

The notion that people aren't 'coerced' to buy conventional electricity, considering all the subventions and tax exemptions and state guarantees and past state spending on R&D for nuclear, hydro, coal or gas, is just ridiculous.

A special note on this misleadingly populistic "people are ready to buy it without being coerced". The obligation an obligation on utilities. The problem for European renewables produces in the early years wasn't that they couldn't find buyers, but that utilities would deny grid access.

Great post. I especially liked the part differentiating the difference in availability of baseload versus renewables. When a coal baseload plant goes down unexpectedly, there's a lot of power that needs to come from someplace. While wind/solar may have a lower capacity factor, they don't tend to have unexpected GW sized drops in generation. The only nitpick I can think of was that as of 2009 wind was ~2% of net generation in the U.S.

Sorry, checking, indeed the 1.33% figure I calculated for wind power as percentage of total consumption including losses (net generation+imports-exports) from EIA annual data (55.363 TWh/4,152.326 TWh) was for the year 2008. As percentage of net generation (4,119.388 TWh): 1.34%. From the monthly data, which is more up to date but I can't find export/import figures, in 2009, wind and net generation was 70.731 resp. 3,953.111 TWh, that's 1.79%.

The precise figures for 2009 German photovoltaic electricity the above percentages are a comparison for: 6.2 TWh/582.5 TWh = 1.06% (as percentage of consumption) resp. 6.2/596.8 = 1.04% (as percentage of net generation).

As usual when this sort of subject is discussed, hydro storage is lightly noted, then largely dismissed. Pumped hydro is very well proven and can be used on seacoasts and caverns as well as in mountains.

Pumped hydro does not have the compression temperature rise energy loss of compressed air.

My favorite save-the-world power is solar thermal on the gulf of california and on the red sea, both of which have the strongest solar flux with a more-or-less infinite supply of seawater to pump up and down. 24 hr power, HVDC to everywhere.

BTW, why does solar thermal take "a lot" of land when the above discussion implies that PV apparently doesn't? Solar thermal takes less land, and it can be almost as small/unit as PV. Also, you can run a solar thermal power plant 24 hrs/day on combustion boosted solar. Makes for much lower $/watt.

"BTW, why does solar thermal take "a lot" of land when the above discussion implies that PV apparently doesn't? Solar thermal takes less land, and it can be almost as small/unit as PV. Also, you can run a solar thermal power plant 24 hrs/day on combustion boosted solar. Makes for much lower $/watt."

Is it because PV can be roof mounted whereas Solar Thermal for power generation can not?

Do you have an estimate for the potential of pumped hydro in any geographical area?

Regarding electricity generating solar thermal and photovoltaic: whether parabolic throughs or towers, the former necessarily needs ground-level areas. Photovoltaic, however, can be, and typically is installed on rooftops (the exception to the rule was Spain's 2008 boom), not taking extra land. (Heat-generating solar thermal is another thing; I myself have a rooftop one that provides 95% of my hot water needs from spring to autumn.) Electricity generating solar thermal is ideal in arid and sparsely populated places more to the South, like Spain or the US Southwest, but it has lots of problems compared to photovoltaic in places like Germany or New England in the US.


Yes, we do have in Spain accurate data. In 2009 2,750 MW of installed power. Some 2.8 percent of all Spanish electric installed power, with an overall efficiency of 0.7 (100 units of energy to pump up -3,703 GWh- and some 70 when generating back.

But most of the pump up installed power was built as a backup the need of some evacuation of 8 nuclear power plants totaling 7,716 MW and generating 52,765 Gwh in 2009 of a total national demand of 266,874 GWh.
There are plans indicating that we should go to some 6,500 MW of specific pump up installed power to be able to cope with the intermittencies of the renewables, expected to cover some 40 percent of the 2020 expected demand (152,835 GWh) for that date.

Hydro is today in Spain 16,657 MW generating 23,236 GWh and cannot grow, as more than 90 percent of the big river basins are already occupied.
I hope this data helps to see the limits to some renewable technologies
What I do not understand is your comments about rooftop PV installations versus on the ground installations. Spain is not the only exception. Italy is following that. And specially, I do not quite well understand why you mention that on the ground installations in Spain have lots of problems compared to rooftop systems in Germany. Apart of having some 30 percent more irradiance in average in Spain than in Germany, massive on the ground installations have much better access, much cheaper maintenance in everything, including logistics of spares, storerooms, control rooms, etc. Much more efficiency in 100 or 500 kW inverters than those 3 to 15 kW scattered domestic toys and a number of etceteras. Could you please sustantiate the mentioned problems, other than aesthetic installations? Because even in the rooftop installations, there are many problems of orientation and shadowing with neighboring buildings than in selected open field installation.

First regarding what I call the 2008 Spanish exception to the rule: in 2008 in Spain, just the photovoltaic farms 2 MW or bigger (see global list here) added up to almost 2.1 GW out of the 2.7 GW newly installed that year: almost the entire capacity was large greenfield plants.

In contrast, more than half of the 3.8 GW of new capacity Germany added last year was plants below 40 kW. Italy is not much different: Italy Surpasses US in Solar PV | Renewable Energy World

According to Gruppo Imprese Fotovoltaiche Italiane (GIFI), 93% of all solar PV in Italy is installed on rooftops in distributed applications. Data from Gestore dei Servizi Energetici indicates that about one-fourth of all Italian solar PV installations are less than 20 kilowatts (kW) in size, or about 300 MW.
  • <3 kW: 6%
  • >3 kW<20 kW: 21%
  • >20 kW<200 kW: 23%
  • >200 kW<1,000 kW: 36%
  • >1,000 kW: 14%

Back to Spain, post-bubble-burst, there is a re-orientation towards rooftop, too:

Spanish PV After the Crash | Renewable Energy World

There is also more emphasis on household systems. Prior to the crash, vast and somewhat controversial ground-mounted arrays made up the bulk of installations. These may now be a thing of the past.

“The current support scheme is better for rooftop and domestic systems than the former one. We are now trying to introduce net metering to the support scheme and the government likes the idea,” wrote Diaz.

Indeed, in a sunny country like Spain, a FIT of €0.32 is enough to make an installation affordable to households and up-to-date figures from ASIF suggest 2010 will see some 600 MW installed. Similarly, the European Photovoltaic Industry Association estimates that this market could continue to add around 375–500 MW a year until 2013, which would keep Spain as one of the top global markets, and enable PV to generate 4%–4.5% of the national electricity demand (equating to roughly 20% of domestic household electricity demand).

Forgot to thank for the hydro data!

Second, you completely misread me about the "lots of problems". I was speaking about solar thermal, not PV farms, and I said Spain is good for them, Germany isn't: (1) most of Germany is either high rain rate arable land or built-up area, so large mirror fields aren't the best use of land; (2) while PV will generate power even when the sky is cloudy, solar thermal efficiency will break down; (3) less Sun due to weather and low Sun in the winter also reduces the potential to provide reliable and regulated power using energy storage. So, while PV will merely give less yield in Germany, solar thermal for electricity generation has much more problems.

You are welcome DoDo.

Then, I did not get your message. Certainly, the solar PV plants on the ground work much better and efficiently than on rooftop mode. Certinaly, solar thermal are still in the learning curve.

Spain has recently passed the US in fed-in installed CSP power and is going up fast, but still solving technical issues and working with materials, but it is miles away from producing as much energy as the solar PV, that in Spain has passed the 2.1 percent of the total yearly averaged national demand.

And with the combined effect of the economic crisis, with a severe drop in the electric demand in one side and the regulations to give access priority into the grid to the renewables, it is fast pointing to some 3 percent.

As for the size of the plants, it depends where you get the figures from. Legally speaking, most of the 3.7 GWp installed up to 2009 in Spain were in plants of 100 kWn or less; in fact, most of them effectively grouped in farms or parks of more than 2 MW; the target for big financial investors. But Italy, has a similar approach of trying to divide into 1 MW plants or even smaller legal units, but also grouped in big farms or parks.

In Germany almost everybody goes to the roof to install. It is probably because they have more economic/energy surplus and even better premium tariffs than Spain and, as you correctly mention, less free lands for their 80 million inhabitants, in a nation of more or less the size of Spain with 46 millions. But efficiencies are much poorer and maintainability as well.

However, diffuse light in cloudy days may give some energy in PV plants, but certainly with dark grey clouds like frequently happens in central Europe, not too much. Even celar greys lower considerably the production.

Some rooftop applications in Italy and France are those considered in Spain “fake” rooftop installations; that is, roofs are first created on purpose (i.e. on greenhouse rural applications that in fact leave only room to grow underneath mushrooms; or stables or warehouses) and then they install the modules. A way to corner legal aspect of on the ground installations.
And back to Spain, I would not trust completely what is mentioned in some articles. It is true that on the roof has received some more attention, only due to the fact that the premium tariffs were kept in the 34-32 c€/kWh while those on the ground are now in the verge of 27 c€/kWh and going down. However, important legal considerations for long term installations are creating doubts in investors. The rooftop installations were mostly oriented to big roofs (malls, factories, commercial centers), than to small private dwellings, who have much less purchasing power and social interest than disciplined Germans. And the problem when there is a deep economic crisis is that the space required to be rented for 25 years extendable periods, can be much better granted in on the ground rural installations, than in factories that tomorrow may have rats as landlords.

Finally, as for ASIF figures, we shall see. For the time being the most accurate and reliable report is the official one of CNE. And in May there were new 230 MW installed in 2010. Let us cross the fingers, because the fossil fueled society is fading and is this society that is paying the premium tariffs (everywhere, not only in Spain) that enables the solar PV development.

Do you have a link for CNE's stats? Would be a good reference for the future. If 230 MW were installed by May, and if PV investors in Spain have the same thinking as everywhere (trying to finish projects before degression reduces rates, that is the end of the year), then methinks ASIF's expectations of 600 MW for the full year are even conservative!

Regarding the size of PV plants in Spain, I don't think that there has to be a conflict between sources. Even if the number of plants smaller than 100 kW well exceeds that of bigger plants, most generating capacity can be in the few big plants (the difference between mean and median), and then that's where most of the feed-in law supported income will be generated. In Italy, where it appears mid-range (100 kW or so) plants dominate, those are still dominantly rooftop (the article I quoted upthread says 93% is rooftop), albeit not private home rooftop, but big rooftop which you mention too. I note that big rooftop was about a third of German installations, too; and a significant part of the larger private household installations is on the roofs of standalone farms, where shadows are less likely to be a yield-reducing issue. (Indeed from the annual stats, taking into account that most of the new capacity coming online each year is concentrated towards the end of the year, I estimate an overall capacity factor around 9-9.5% last year and 10% the year before, which is not much below what one can expect ideally in Germany, 11%.)

Regarding long time ranges and being target for big financial investors, I see that with more sceptical eyes. When Spain had its big boom in 2008, it was because the difference between the feed-in rate (then identical for on-the-ground and rooftop installations) and falling product prices became so big that much quicker return on investment was possible, drawing in venture capitalists. In other words, I would trust such an investor, who cashes in in the first 5 years, much less to shoulder the maintenance costs over the notional 20-25 year lifespan of a solar panel than a homeowner (or a farmer, school, public institution), which counterweighs the maintability disadvantage in my eyes. (There are parallels in the pioneering years of wind power in the USA/California, with reckless investors rushing in to cash in on subventions and tax breaks with farms made of cheap but substandard turbines, and then abandon those after the first breakdown.)

A further issue, which I don't have data on but maybe you do, is grid access: what was the cost factor of connecting MW-scale solar farms to the grid (resp. improving the capacity of connecting lines) in Spain? This is interesting because in the comparison with rooftop solar (at least smaller scale), grid connection is more or less a given (though from the investor rush it appears the benefits still favoured on-ground big PV).

Please note the link to the public CNE data
It is a ZIP excel spreadsheet updated in monthly or bi-monthly basis

The figures in a given period f the year are not to take from granted inter-annual extrapolations. The installations to be effected will be those officially preregistered and they are more in the range you mentioned before, somewhere below the 500 MW/year, if the economy still allows it. And as you mention, the accelerations of cutovers at the end of the year, have sometimes financial or fiscal reasons, rather than technological ones.

As for the size of plants I believe that the important issue is to verify, as CNE does, the final production really invoiced at national level. This is what it counts and is the most evident approach to the truth.
The long time ranges are needed by the big fishes, because they usually represent pension funds behind and need to guarantee stable and guaranteed (in this case by Royal Decrees) inputs throughout a long period of time. This is independently that they may get their own management fee sooner than later. In this complex financial world, there is also a second type of fishes, the family houses or investors happy with ROI’s with one digit. But in all the cases, now in Spain, you do not get a single Euro from the bank to leverage an operation, if there is not on the table before a complete technical and legal due diligence, stating that the plant will perform well for the next 15 to 25 years. And of course, they are still fallible.

And for the data on grid access, yes, I have plenty of data and I expect to release a book later this summer on this aspect, something never included properly in the BoS of the PV LCA’s or EPBT’s. However, you should not underestimate or extrapolate that rooftop domestic installations will have advantages or will not require extra fed-in or infrastructure or specific control and remote management devices, versus the big on the ground installations. Perhaps somehow in the contrary. Any energy coming from the extreme of highest capillarity of an electric network (bottom up) like solar PV creates a big problem to manage the present networks, when it reaches a certain level. For instance, one of the problems encountered by network managers in Spain is a number of small to medium size PV plants that were fed in connected without the remote siwtch off/on equipment, so that some labor force when maintaining lines, they may be electrocuted, if when starting measures in the maintenance point the PV plant was at zero level (i.e. clouds, and when in operation, the sun starts shinning).

If certain capacity is added to a given line, the whole line has to be changed, be that offered from a big plant of from a number of scattered plants. The regulatiosn says that for a 10 MW capacity line in a classic, conventional top-down flow, the bottom up energy can not exceed 5 MW (half of the nominal capacity. Therefore, the extra infrastruture needed will be more or less the same at the end of the day and when thinking in something else that few private entrepreneurs

Many thanks for the link and the points on grid connection. But isn't the last paragraph unfinished?

However, such demand-side solutions tend to have their limitations due to seasonal variations: for example, electric heating is not used in the summer.

But hot water heating is always used in the summer and flexible freezers, fridges, cold warehouses etc. are also always used in the summer and so is air conditioning in many places, which can also be operated flexibly in a well insulated building and/or in combination with water/ice storage.
But more importantly: Wind energy output is always lower in the summer in Europe - so hopefully less electric energy is needed in the summer.

Unfortunately, people keep on forgetting that significantly more energy is needed even just for hot water than for any non-heat-energy-related (inflexible) electric appliances:

In any case, it won't make sense to, for instance, stop air-travel just to be able to keep oil powered water heaters and oil powered heating systems running.

I am very glad to see all this good discussion. I wish to add a few points that most people don't even know about.

Solar thermal on rooftops. Quite possible, using stirling engines instead of steam. A mini-power tower with many small heliostats aiming at one, say, 100kW stirling , all mounted on top of a Wallmart in LA, is the sort of concept I had in mind. Certainly no good at all in Hamburg, but not everywhere is gloomy.

Where do we get that 100kW stirling? No problem. We get it from the good people who finally see the light (ha!) and grab the business opportunity from what is already in the tech literature and turn it into hardware that people can buy.

And on hot water- it seems absolutely crazy that in the US, even in entropy pits like the place where I live, it is EASY to get all the hot water from sun in at least 6 months of the year. I do it, anybody can do it.

Overall good post. Has problems, but pretty good.

The redefinition of the 3 different generation categories is simply a truism. But those who defined the 3 groups by how they looked on the graph never really understood subject in the first place.

And the drawing of the graph (first graph in post) shows that the author has not reached these conclusions looking at the actual data. The unplanned disconnects due to baseload depicted there is monumentally exaggerated from reality. In one full day, in a large power service region, the most likely case is that baseload plants will not encounter random shutdowns.

Nuclear metric:
0.6 unplanned SCRAMS per 7,000 hours critical.

Sorry for the industry jargon, but this basically says that nuclear plants trip off the grid less than once per year. I'll go ahead and admit the bias of the metric - if you look at operating history of these plants, they go offline more often, but those aren't unplanned and unexpected. Nonetheless, the fact that the largest of baseload plants don't trip off once per year throws a huge wrench in your argument and makes me question the aptitude of those who stress the "unreliability" of baseload power.

There are other problems with baseload power, and I am a fan of the definition of it in this post, as well as the inclusion of wind in it.

I'll also express agreement with the need for more intermediate sources in the future. That is very very true. We have a problem creeping up on us, and it's been creeping up for the last 2 decades as well. The matter of 24/7/365 sufficiency and robustness in grid operation will be a major technical challenge as we further advance into the post-industrial age.

Thanks for advancing the discussion.
But be careful with your claims in the future.

0.6 unplanned SCRAMS per 7,000 hours critical.

Would you happen to have a source for that? Is that US, or elsewhere?

Nonetheless, the fact that the largest of baseload plants don't trip off once per year throws a huge wrench in your argument and makes me question the aptitude of those who stress the "unreliability" of baseload power.

No one is suggesting that nuclear is not "reliable enough", or that it's production variance isn't much smaller than wind & solar. What they are suggesting (or, at least I am suggesting) is that nuclear, along with all other sources of generation, have their own "intermittency" which is significant and which must be planned for: scrams disconnect very large (1GW) amounts of power in minutes, which presents a very, very large rate of change. For instance, Ireland has decided against nuclear because the loss of a single 1GW plant would destabilize their small grid.

The nuclear community is acutely aware of the challenges of large capacity additions, and the push to produced "right-sized reactors" is a fairly direct response to that issues. And yes - we're not there yet.

1 GW is perhaps an understatement. Current designs being marketed go to 1750 MWe. They could go higher. What's more, these units don't participate in automatic grid control, thereby diluting any stabilizing frequency response that the other units offered.

For source on scrams, the NRC actually sets limits on unplanned SCRAMS:

And the 0.6 number can be found from industry sources, WANO:
(this is a worldwide statistic, the USA should be better, but I don't know, if you dug deep enough into NRC data I'm sure you could figure it out)

But let's just get clear about what kind of beast a 1 GW (1.8?) loss on the grid is. That is a low probability, high significance event. The planning for nuclear capacity loss is a matter contingency planning. On a national scale (and on the scale of FERC region), the fleet of nuclear and large coal and combined cycle plants are very reliable. Similar to how people talk about the benefits of connecting many wind farms - though individual units are intermittent, the entire fleet can scarcely be called intermittent (I mean nuclear & coal baseload, not wind).

I agree with your description of contingency planning for nuclear capacity loss.

I like to emphasize the similarity between nuclear and wind that you discussed. On an individual plant basis, nuclear has an important level of variance (as do all other forms of generation). On a continental basis, that variance is extremely manageable. While wind obviously has more variance than nuclear, the overall situation is pretty similar: on a continental basis, wind variance becomes pretty manageable.

Thanks, yes, the point is intermittency, not reliability (and not just nuclear but coal too).

The 8 nuclear power plants existing in Spain, with a total installed power of 7,716 MW produced in 2009 52,765 GWh. That gives a total of 78 percent load factor. This represents 6,872 hours per year, close to your 7,000 hours critical. Perhaps because in Spain last year there were probably some more than the 0.6 unplanned SCRAMS with an installed park about 30 years old.

However, the consideration of 24/7/365 sufficiency of the nuclear power plants is considering that business as usual goes as usual. The long supply chain from the mining of the original fuel, in less countries than those today producing oil, or gas or coal; the limited amount of proven reserves; the announcements of many countries seeking for more plants, thus dramatically reducing the proven reserves/extraction ratios, if finally successful.

And over all, the many concerns and security considerations for countries, if they can not freely enrich the fuel, or have and exert full control of all the supply chain, for lack of credibility of the so called "International Community" in their final intentions, does not permit them to be so sure in the long term about the "sufficiency" and "robustness", if any of the many parts of the supply chain (U235, fuel rods, complex piece parts, reprocessing of wasted material, etc., etc.) in alien hands fail, for whatever the reason, from embargoes to blockades or commercial sanctions.

Not to speak of wars like the bombing of the Iraqi nuclear power plant Osirak by Israelis in early eighties, or what is in the air now with the nuclear power plant in Iran.

What would have happened if the Sha had lasted few years more, enough to complete the 20 nuclear power plants happily approved and already programmed by Western nations and Khomeini had arrived to power later, when in operation? Shall we be in a position to state today that these plants will continue today an example of "sufficiency and robustness"?

Since Iran ruled by Khomeini managed to maintain F-14 fighters that are tempramental maintainance hogs I expect they in a alternative history universe would have managed to keep 15-18 plants out of 20 running and would thus would have had a much stronger economy.

The Persian culture seems to be good with engineering but they have made a mess of their political development and that were made worse by other countries meddling and the Irak - Iran war that were about as bloody as the first world war. This is no excuse since people can overcome such things and build their own greatness instead of focusing on hating other people.

@Pedro Prieto:
the limited amount of proven reserves; the announcements of many countries seeking for more plants, thus dramatically reducing the proven reserves/extraction ratios, if finally successful.

Indeed what makes me most sceptical regarding a big global expansion of nuclear is the question of supply. The reduced time for which proven reserves would remain sufficient is just the first of the problems I see. The thing is that currently, reserves of the higest grade are tapped, even among the reserves that can be mined economically at current uranium prices (not to mention granite or ocean water). Going for lower grades would mean (1) a dramatic increase of mining volume, to scales similar to that of coal mining (with connected environmental destruction and potential conflicts with locals), (2) a Peak Oil-style problem.

About the latter: When Peak Oil sceptics bring up oil shales, they forget that the issue is the speed of extraction, not the size of reserves. Similarly, for lower-grade uranium reserves, the pressing issue is not their size, but how fast and to what level the mining industry can run up their extraction.

First of all, those diagrams are for the illustration of phenomenons (the interactions of base, intermediate and peak load), they are not an attempt at a reproduction of reality. (I can also reassure you that I saw plenty of actual load curves.) But, even having said that:

  • The power scale of unplanned disconnects relative to the full power depends on the size of the country and the outage. A single nuclear reactor outage would be less of a dent than on my diagram for the US or the biggest EU members, however, would just be right for Belgium (7 nuclear blocks giving baseload and more) or Hungary (4 blocks). Furthermore, there is the (much rarer) possibility of a whole plant with multiple blocks going off-line, or the possibility of cascading shutdowns.
  • The time scale is obviously not a single day (neither for the frequency nor the length of outages), but a realistic, at least week-long drawing would have been a less clear illustration.

Regarding realistic baseload downtimes, I note that

  • methinks a metric more relevant than 0.6 unplanned SCRAMS per 7,000 hours critical is the unplanned capability loss factor of 1.6% (which is still not a full measure of accident-related downtime because some scheduled maintenance is scheduled with the aim to mitigate shortcomings discovered in the analysis of an accident);
  • both metrics relate to one plant, not a system;
  • these are global averages.

If you take Germany, with 17 active reactors, of which two are currently down due to accidents and have been for three years (save for a few hours between two accidents for one of those plants), the picture is less rosy. (BTW, the shutdowns two hours apart three years ago were most likely an example of a cascade.)

In addition, baseload doesn't consist of nuclear only. This US study (h/t to Gail) investigates what constant power the combinations of 19 existing US wind farms spread over an area of 850 km x 850 km can provide with the same reliability as the average of US coal plants. They quote five-year averages of scheduled shutdowns 6.5% of the year and unscheduled outages 6% of the year.